Section 4: Subspecialty Management
Chapter 59: Pediatric Anesthesia

Neonatal Anesthesia

The neonate has unique requirements for equipment, intravenous access, fluid and drug therapy, anesthetic dosage, and environmental control (Ch. 58). The many neonatal procedures are too numerous to describe in this limited space. However, the basic anesthetic management is the same for all neonates. An understanding of the basic differences in physiology, pharmacologic and pharmacodynamic response, and the underlying pathology of the surgical problem is essential for development of a safe anesthesia plan. Most of the complications that arise are attributable to a lack of understanding of these special considerations prior to induction of anesthesia. The care of neonates is fraught with danger, sudden changes, unexpected responses, and the unknown congenital problem. If anesthesiologists are to deliver optimum pediatric anesthesia care, they must always be prepared for the unexpected, have the proper size and variety of equipment available, and obtain the highest level of support, both in the operating room and in the intensive care unit.

Children younger than 1 year have a higher incidence of complications than older children. 197, 198, 199, 200, 212, 213  These complications relate to oxygenation, ventilation, airway management, and response to anesthetic agents and medications; they occur more frequently in ASA physical status III and IV patients. The neonate, particularly the premature infant, functions on a marginal basis, so that any type of stress is usually poorly tolerated. This vulnerability may relate to the technical difficulty of taking care of small patients, the immaturity of their organ systems (especially the cardiovascular, pulmonary, renal, hepatic, and nervous systems), the high metabolic rate, the large ratio of body surface area to weight, and the ease of miscalculating a drug dose.

When caring for infants and neonates, special attention must be paid to all aspects of anesthesia and surgical management. The anesthesiologist must devote particular care to the calculation of drug dosage and the dilution of drugs. Prevention of paradoxic air emboli requires that all air be vented from intravenous devices and syringes prior to use (aspiration of each intravenous injection port to remove air trapped at these junctions). A volume of air that is clinically unimportant to the adult may prove catastrophic to the infant. Warming of preparation solutions and irrigation fluids prior to application minimizes heat loss. Intravenous fluids should be administered with volume-limiting devices; infusion pumps are particularly helpful in preventing overadministration of intravenous fluid. The composition and infusion rate of flush solutions should be noted and calculated into maintenance fluid therapy.

Every effort must be made to maintain the infant‘s temperature to minimize thermal stress. The operating room environment should be warmed so that the whole operating room constitutes a giant incubator. In addition, the infant weighing up to 10 kg benefits from a warming blanket, whereas most patients benefit from heated humidification of inspired gases, particularly when a nonrebreathing circuit is used. Heated air mattress devices are particularly useful for maintaining temperature.

Monitoring of expired concentrations of carbon dioxide will be less accurate if the sample is taken at the Y-connector; endotracheal tubes having a built-in sampling port may improve the accuracy of monitoring. Pulse oximetry is definitely of great value, not only in diagnosing hypoxemia, but also in preventing extreme hyperoxia. Maintaining oxygen saturation at 93 to 95 percent keeps the preterm infant on the steep side of the oxygen-hemoglobin dissociation curve. This consideration is important to infants still susceptible to retinopathy of prematurity, that is, those younger than 44 weeks‘ PCA. Because these infants have the highest oxygen consumption, an oxygen saturation in the 93 to 95 percent range can result in hypoxemia within seconds. When managing such a delicate balance, and bearing in mind the slight inaccuracies of these monitors, 214  the anesthesiologist must be extremely vigilant and prepared to respond rapidly to changes in oxygen saturation. Keeping arterial carbon dioxide values within the normal range (35–45 mm Hg) may also be important in preventing retinopathy of prematurity. 215  However, many other factors beyond the control of the anesthesiologist often contribute to the development of this condition. 216, 217 

The Stress Response

There is little doubt that the neonate, even the extremely premature infant, is capable of feeling pain and responding to painful stimuli; what actually constitutes a state of “anesthesia” has yet to be proved. 218, 219  No child should be denied analgesia or anesthesia because of size or age. Clearly, the cardiovascular system of the premature infant rarely tolerates the cardiovascular-depressant effects of potent inhaled anesthetics. However, the narcotics (e.g., fentanyl, sufentanil, alfentanil, remifentanil) are usually well tolerated even by critically ill infants. These potent narcotics must be carefully titrated to response, and the anesthesiologist must always be wary of narcotic-induced bradycardia and its consequences on cardiac output. Low concentrations of potent inhaled anesthetics can be used with narcotics, thus providing a way of controlling hemodynamic responses without significantly depressing the myocardium. The relative merits of one anesthetic technique over another are not clear, and the few studies examining this issue are poorly controlled. 220  Narcotics and inhaled anesthetics suppress the hormonal responses to pain. 221  The important result of the provocative studies undertaken to clarify this issue is that they have heightened everyone‘s awareness of the need to provide adequate anesthesia and analgesia for the neonate.

Special Problems


Meningomyelocele (hernial protrusion of a part of the meninges and spinal cord through a defect in the vertebral column) is a relatively common neonatal abnormality that requires surgery. The following should be considered in addition to the usual concerns for management of the neonate: (1) the possibility of underestimating fluid and blood loss from the defect; (2) the high association of this condition with hydrocephalus; (3) the possibility of cranial nerve palsy, resulting in inspiratory stridor; and (4) the potential for brain-stem herniation. The anesthesiologist must establish adequate intravenous access and invasive monitoring if appropriate, replace all fluid deficits, including loss from the defect (usually with normal saline), and ensure that crossmatched blood is available. Latex allergy precautions should be used with these patients for their first and all subsequent anesthetics. 222 

Pyloric Stenosis

Pyloric stenosis normally presents in the first 3 to 6 weeks of life. The following are major concerns for the anesthesiologist: (1) a full stomach, often filled with barium sulfate; (2) metabolic alkalosis with hypochloremia and hypokalemia; and (3) severe dehydration.

This operation is never a surgical emergency. The patients should be carefully evaluated, and severe metabolic imbalance should be corrected prior to surgery. The stomach should be suctioned with a wide-bore catheter in the supine, right, and left lateral positions immediately before induction of anesthesia, and as much barium as possible should be removed. 223, 224  Patients with pyloric stenosis can be managed with awake endotracheal intubation or rapid-sequence induction of anesthesia. Arguments can be made for either type of induction, and selection depends on the skill and familiarity of the anesthesiologist with that technique.

Omphalocele and Gastroschisis

Omphalocele and gastroschisis are major defects in closure of the abdominal wall, resulting in exposure of viscera that are either covered (omphalocele) or not covered (gastroschisis) by peritoneum. The major problems with these defects include the following: (1) severe dehydration and massive fluid loss, both from the exposed visceral surfaces and from third-space losses caused by partial bowel obstruction; (2) heat loss; (3) the difficulty of surgical closure; and (4) the high association of this condition with prematurity and other congenital defects, including cardiac abnormalities.

These patients must be cared for expeditiously to minimize the potential for infection and the compromise of bowel function and to reverse the loss of fluid and heat. Adequate intravenous access and invasive monitoring are usually necessary. The liberal use of muscle relaxants provides optimal surgical conditions for closure of the defect; hypotension secondary to tension on a major organ (liver) or caval compression is common. If the surgeon is unable to close the defect in one procedure, a staged procedure is planned. In either situation, postoperative ventilation is necessary until the abdominal wall has had time to stretch to accommodate the viscera. Intravenous alimentation may also play a vital role in the rapid recovery of these patients. A small percentage of patients with omphalocele also have Beckwith-Wiedemann syndrome, a condition characterized by profound hypoglycemia, hyperviscosity syndrome, and associated visceromegaly. It should be noted that increased abdominal pressure following closure may compromise hepatic function and may markedly alter drug metabolism. 128, 129 

Tracheoesophageal Fistula Anomaly

The tracheoesophageal fistula anomaly has five or more configurations, most of which present as an inability to swallow because of esophageal atresia (the ending of the esophagus in a blind pouch). Neonates may have aspiration pneumonitis from a distal fistula connecting the stomach to the trachea through the esophagus or from a proximal connection of the esophagus with the trachea. The characteristic diagnostic test is the inability to pass a suction catheter into the stomach. Additionally, this anomaly may be a part of a larger constellation of anomalies known as the VATER syndrome (V, vertebral; A, anal; TE, tracheoesophageal; R, renal). Any patient who presents with tracheoesophageal fistula or esophageal atresia should be suspected of having the other anomalies described earlier, as well as the potential for congenital heart disease.

The major issues for management of safe anesthesia include the following: aspiration pneumonia; overdistention of the stomach because of entry of air directly into the stomach through the fistula; inability to ventilate the patient because of the large size of the fistula; problems associated with other anomalies, particularly a patent ductus arteriosus (shunting); and postoperative intensive care.

For management of anesthesia, the child should not receive any feedings, should have a catheter placed in the esophagus to drain saliva, and should be placed prone in a head-up position. If the child has pneumonia, treatment should be initiated; surgery may be postponed until pneumonitis improves or clears. The child may be a candidate for gastrostomy to provide a means of nutrition during recovery from pneumonitis. Anesthesia evaluation should center around the pulmonary and cardiovascular systems. Generally, an awake intubation is performed, and the endotracheal tube is intentionally passed into the right main stem bronchus; the endotracheal tube is then slowly withdrawn until breath sounds are heard on the left. Often this technique ensures that the tip of the endotracheal tube is placed beyond the origin of the fistula, thus avoiding massive distention of the stomach. Care must be taken to avoid rupturing the stomach; therefore, spontaneous and gently assisted ventilation may be appropriate until gastrostomy is performed. A change in the distance of insertion of the endotracheal tube of as little as 1 to 2 mm may determine whether the anesthesiologist is ventilating both lungs, one lung, or the fistula. Therefore, because a change in oxygen saturation may be the first indication that all is not well, the pulse oximeter is one of the most useful monitors in managing these patients. 225  Taping the stethoscope to the left side of the chest in the axilla also decreases the possibility of unrecognized endobronchial intubation.

Diaphragmatic Hernia

Infants with diaphragmatic hernia usually present on the first day of life with respiratory distress and a scaphoid abdomen. The abnormality is a herniation of the abdominal viscera through a defect in the diaphragm, most commonly the foramen of Bochdalek on the left side. Almost all the abdominal viscera, including the liver and spleen, may be above the diaphragm. The GA at which herniation occurs may determine the degree of lung hypoplasia.

Anesthesia concerns include hypoxemia and hypotension caused by overdistention of the stomach and herniation across the midline, hypoxemia because of primary pulmonary hypoplasia, hypoxemia due to pulmonary hypertension, and systemic hypotension caused by kinking of major blood vessels, particularly those of the liver. In general, the anesthesiologist‘s ability to control arterial carbon dioxide tension (PaCO2) reflects the severity of lung pathology and therefore survival. An inability to reduce PaCO2 is associated with poor prognosis. 226  Extracorporeal membrane oxygenation has reduced the mortality associated with this condition. In addition, the urgency for surgical intervention has diminished, giving way to a desire to stabilize the infant and minimize stress. 227  At some centers, the common practice is to postpone surgery on infants with diaphragmatic hernia until their condition has been stabilized for several days.

Anesthesia management for patients with diaphragmatic hernia includes the following:

  1. An awake endotracheal intubation without bag-and-mask–assisted ventilation prevents overdistention of the stomach and herniation across the midline.

  2. Insertion of an arterial line and close observation of the surgical field are most helpful in diagnosing impairment of venous return or cardiac output.

  3. Blunting of the stress response is accomplished by providing analgesia with narcotics (usually fentanyl) and by controlling respiration with muscle relaxants (usually pancuronium).

  4. Careful control of ventilation and oxygenation prevents sudden increases in pulmonary artery pressure (PaCO2 is maintained at less than 40 mm Hg and PaO2 at more than 100 mm Hg). Pulse oximetry is helpful in diagnosing subclinical episodes of hypoxemia.

  5. Hypothermia is avoided in order to decrease the oxygen consumption needed for thermogenesis.

  6. Anesthetic agents that could depress the myocardium are avoided until the chest is decompressed.

  7. To prevent bowel distention, nitrous oxide is not given.

  8. The anesthesiologist should be alert to the development of a barotrauma-induced pneumothorax on the ipsilateral or contralateral side.

  9. The anesthesiologist should ensure adequate intravenous access for maintenance of a constant circulating blood volume.

  10. Postoperative intensive care should be planned.

The Former Preterm Infant

Since the early retrospective reports noting the high incidence of postoperative apnea in former preterm infants (GA <37 wk), several studies have tried to define the population at risk 187, 228, 229, 230, 231, 232, 233, 234, 235, 236  (Ch. 58). Most studies have found the majority of infants who develop postanesthesia apnea to be less than 46 weeks‘ PCA; however, apnea has been reported in infants up to 60 weeks‘ PCA. 231  The conclusions reached by each of these studies were limited by the relatively small number of patients. I had the opportunity to obtain and analyze the original data from eight prospective studies. 187, 228, 229, 230, 231, 234, 236, 237  A combined analysis examining risk for apnea included only patients undergoing inguinal hernia repair and not receiving special treatment such as caffeine or regional anesthesia. 188  The following risk factors were examined: a history of respiratory distress syndrome, bronchopulmonary dysplasia, neonatal apnea, necrotizing enterocolitis, ongoing apnea at the time of surgery, use of narcotics or long-acting muscle relaxants, anemia (hematocrit level <30%), GA, and PCA. The only two risk factors that stood out across all ages were GA and PCA. The incidence of apnea was inversely related to both GA and PCA. For example, if two infants were now 42 weeks‘ PCA yet one was born at 28 and the other at 36 weeks‘ GA, the 28-week-GA infant would have approximately twice the potential for apnea (Fig. 59–12). The incidence of apnea was higher in institutions that collected data with continuous recording devices compared with those that relied on impedance pneumography or nursing observations. Anemia (hematocrit level <30%) is an independent risk factor associated with apnea in former preterm infants (see Fig. 59–12). What makes this risk factor unique is that the risk of apnea in anemic former preterm infants is not altered by GA or PCA; that is, the risk of apnea would appear to be the same for an anemic 60-week-PCA infant as for an anemic 45-week-PCA infant. Even after eliminating infants with an obvious apnea spell in recovery and infants who were anemic, the risk for apnea does not decrease to less than 1 percent with 95 percent statistical confidence until PCA of 56 weeks with GA of 32 weeks or PCA of 54 weeks with GA of 35 weeks. It would therefore seem prudent to admit all former preterm infants younger than 55 weeks‘ PCA to monitored beds and all anemic former preterm infants.

Click thumbnail to see full size image
FIGURE 59–12 Predicted probability of apnea for all patients, by gestational age and weeks of postconceptual age. Patients with anemia are represented by the horizontal hatched line. Bottom marks indicate the number of data points by postconceptual age. The risk for apnea diminishes for infants born at a later gestational age. The shaded boxes represent the overall rates of apnea for infants within that gestational age range. The probability of apnea was the same regardless of postconceptual age or gestational age for infants with anemia (horizontal hatched line). (From Coté et al188 )

Although the risk of apnea may be less with regional anesthesia, apnea may still occur and may even be increased if regional anesthesia is combined with sedation (ketamine, midazolam). 230, 235, 238, 239  There are insufficient numbers of well-controlled studies of regional versus general anesthesia to determine whether the risk is significantly reduced.

Part of the problem of how to treat these children is that we do not know the clinical importance of brief apnea/bradycardia spells that do not require intervention. Because cerebral blood flow is markedly reduced with heart rates lower than 80 beats/min, even brief apnea associated with bradycardia may have adverse effects. 240  The problem faced by the clinician is how to manage former preterm infants scheduled for outpatient procedures. Information concerning the true incidence of postoperative apnea is still very limited because the available data have been collected from so many institutions. Nevertheless, one can conclude that the incidence of apnea is inversely related to PCA and GA, and anemia is also an independent risk factor. 188  More prospective studies need to address whether regional techniques are safer than general anesthesia. Studies must also include methods (nasal thermistry or capnography) for detecting obstructed respiration, which is often missed by chest wall impedance monitoring. Pulse oximetry would help to define the severity of desaturation, and ECG monitoring would help define episodes of bradycardia, which may reduce cerebral blood flow. Kurth and LeBard 234  demonstrated that up to 12 percent of apnea spells are accompanied by desaturation to below 80 percent. It would seem reasonable that the combination of desaturation and bradycardia would be of more physiologic importance than simple documentation of pauses in respiration.

Because the available data are so confusing, it would appear that outpatient anesthesia should be considered only when former preterm infants have had a totally unremarkable neonatal history and are currently healthy. If there is any question, the prudent anesthesiologist will plan on postoperative admission and monitoring. If appropriate facilities are not available, the premature infant who is less than 60 weeks‘ PCA should be referred to an institution that has such facilities. High-dose caffeine (10 mg/kg) has been recommended. 229  This may be an effective therapy; however, the half-life of caffeine in older former preterm infants is only 6 hours, 241  and the first apnea spells following anesthesia may not manifest until 12 hours. 188  Therefore, one cannot administer caffeine and send the infant home and assume that the problem is treated.

To make the issue even more complex, I have observed a full-term infant who became apneic following general anesthesia. This infant had periodic breathing, which is distinctly unusual in a full-term infant. 242  We still do not know with certainty which patients are truly anesthetized safely on an outpatient basis and at what PCA and GA.