41 Surgery in neonates is accompanied by a humoral stress response that leads to increased complications and mortality, as shown in landmark studies by Anand and colleagues in 1987.3,4 This response is diminished by adequate anesthesia (Figure 41-1; Table 41-1), resulting in improved surgical outcomes. Consequently over the subsequent decades, anesthesia for all surgical procedures in neonates has been accepted as both a clinical and an ethical imperative. TABLE 41-1 Data from Anand KJ, et al. Randomised trial of fentanyl anaesthesia in preterm babies undergoing surgery: effects on the stress response. Lancet. 1987;1:62. It is also now generally accepted that neonates are capable of sensing pain and discomfort. We have no recallable memories before 3 to 4 years of age (termed infantile amnesia), but neonates do possess consciousness and a sense of self, and can form implicit memories, that is, changes in behavior based on prior experience that do not require intentional recall.16 Plasticity is an inherent characteristic of the developing nervous system, and early pain experiences can lead to exaggerated responses to later painful stimuli or stress, as well as impaired neurodevelopmental outcome.78,79 Changes in pain sensitivity in children exposed to nociceptive stimulation as neonates have been demonstrated using parental rating surveys,33 measurements of the child’s own perception,32 and psychologic testing.11,31,39 Activation of brain regions in response to graded moderate pain stimuli is significantly different in school-age children and adolescents with neonatal intensive care (NICU) experience compared with controls, as measured by functional magnetic resonance imaging (MRI).40 Based on such evidence, the goals of anesthesia in neonates are not restricted to prevention of the stress response to surgery, but also effective management of pain and discomfort.1,2,16 Coincident with a new understanding of the importance of anesthesia and pain management in neonates, since 1999 numerous animal studies, including some in nonhuman primates, have demonstrated abnormal neuroapoptosis and, in some, long-term cognitive defects after early exposure to virtually all commonly used anesthetic and analgesic agents.10,30,42,44,76 Development of the nervous system is complex, involving neuronal proliferation, migration, differentiation, and “pruning” accomplished by apoptosis. This development is in part dependent on the balance of excitatory and inhibitory stimuli and neurotransmitters. Given the complexity of the process, as well as the obvious physiologic and developmental differences inherent in application of animal studies to humans, the overall impact of these studies remains uncertain. Concerns have been raised with respect to the animal data regarding dosing and the time course of exposure. A clinically appropriate dose is the minimum necessary to induce anesthesia in the species. In some species, and with some agents, this dose also results in significant mortality, increasing the difficulty of interpreting the significance. Synaptogenesis in rats occurs postnatally over the first 2 weeks of life. An analogous period in humans would range from the third trimester of pregnancy through the first 3 years of life. Whether the child might be susceptible to anesthetic toxicity during this entire period is unknown. Other studies looking specifically at neurodevelopment suggest that a postnatal 7-day-old rat more closely corresponds to the human fetus between 17 and 22 weeks of gestation. Elucidation of the critical period is of prime importance in anesthetic management.55 Finally, most of the animal studies of apoptosis involve anesthesia without surgery. Because neurodevelopment is related to the balance of stimulatory and inhibitory neurotransmitters, there is some indication that the absence of surgical stress may alter the neurodevelopmental response to anesthetics. At this point, studies in humans are predominantly retrospective cohort studies, some of which suggest a possible association between neonatal anesthetics and longer term neurocognitive defects. There is an ongoing initiative by the US Food and Drug Administration (FDA) to study these risks.60 Confirmation of such long-term and relatively subtle effects in humans is of course a daunting task. At this time, there have been approximately 14 studies of anesthetic effects in humans.5,9,20,21,25,26,34,36,43,46,74,75,84,89 These studies have recently been reviewed by Gleich et al., along with a summary of available animal data.30 Of the human studies, eight demonstrated a potential effect of anesthetic exposure,9,20,43,84 although in four of these studies, it was only after multiple exposures.21,25,74,89 Only one uses prospective data in a limited 1-year follow-up of infants undergoing pyloromyotomy.84 The studies all use varied sources of data, either from a single facility, a single payer system, or a defined geographic area. In many of the studies, there is little information with respect to the specific anesthetic agents and/or techniques used. The outcome measures utilized also vary, including group-administered achievement tests, individually administered achievement tests, parent and teacher ratings, and diagnostic codes of neurodevelopmental disabilities. The fact that significant adverse outcomes were seen only with multiple anesthetic exposures raises the question of the effect of comorbidities on the apparent anesthetic effect. The adverse outcome owing to multiple exposures did persist in one study in which an attempt was made to rigorously control for comorbidity.25 Should these data cause any change in current clinical practice? No one can question the overwhelming benefit of anesthesia in the setting in which a surgical procedure, or any noxious procedure with significant stress, is required. The detrimental effects of surgical stress have been well shown. Similarly the benefits of early repair of some surgical problems, such as tetralogy of Fallot or neonatal hernia with risk of incarceration, have been shown, although such repair may lead to anesthesia at a younger, and potentially more susceptible, age. The human studies of anesthetic outcomes suggest minimal effect with single brief exposures. In addition, as discussed by Gleich et al.,30 a cautious and balanced approach to interpretation of the human epidemiologic studies is warranted given the overwhelming known benefits of anesthesia. As they also point out, this does not mean a lack of concern, as a modest increase in the risk for cognitive problems given the millions of exposures each year would have major implications. There is a clear need for large, prospective studies, with a defined end point, but this is a daunting task, and it will take years to complete. Personnel, equipment, and the operating room environment need to be specifically adapted for neonates. Anesthesia-related morbidity is decreased in children anesthetized by pediatric anesthesiologists compared with children cared for by non–pediatric anesthesiologists.41 Monitors, anesthesia delivery systems, mechanical ventilators, and environmental controls all need to be appropriate for use with neonates. One consequence of the reduced FRC and high metabolic rate in a neonate is a diminished reserve. Changes in the fraction of inspired oxygen (Fio2) are rapidly seen as changes in Po2, and the neonate quickly desaturates if ventilation is interrupted. This situation limits time for intubation, and airway management can be difficult. The high alveolar ventilation also accounts for a very rapid uptake of inhalational anesthetic agents, especially in premature infants, making it easy to overdose with these agents. Closing volume, which is the lung volume at which smaller airways tend to collapse, is very close to FRC in neonates. It is well known that anesthesia causes decreases in FRC.38 In a neonate, this low closing volume can result in airway closure at end expiration, with resultant atelectasis, ventilation/perfusion mismatch, and increased intrapulmonary shunting. Airway anatomy in infants differs from anatomy in older children. The infant’s head is much larger compared with body size than that of older children, although the infant’s neck is short. The infant’s tongue is large, but the larynx is higher and anterior, with the cords located at C4 in the infant compared with C5 or C6 in an adult. The epiglottis of the infant is soft and folded. The neonate’s larynx has commonly been described as conical, with the narrowest point in the subglottic area at the cricoid ring, based on an often-quoted article by Eckenhoff. Eckenhoff’s description was apparently based on earlier measurements from plaster castings of a few cadaveric larynxes, and by his own admission, may not apply to living infants.6,63 Two more articles revisit this anatomy using magnetic resonance imaging (MRI)52 and video-bronchoscopic imaging.15 Based on these studies, the larynx is cylindrical, although not round in cross-section, but rather elliptical, with the anteroposterior dimension slightly greater. A tight-fitting, round endotracheal tube might compress the lateral laryngeal mucosa; subglottic stenosis remains a common complication, especially with longer-term intubation. Although uncuffed endotracheal tubes previously were used in neonates and children, use of newer cuffed tubes, composed of very thin, low pressure cuffs and thin walls, is feasible in many infants, although an uncuffed tube is still required in the smallest neonates. New tubes are available with very thin cuffs, and without the Murphy eye, which decreases the length of the tube below the cuff. This ensures that the entirety of the cuff is below the cords, avoiding pressure on the cords. A cuffed tube can be sized smaller because the cuff prevents leakage and could decrease the incidence of subglottic stenosis, rather than increase it, as previously believed. Laryngeal and tracheal trauma is important because even modest airway edema can be serious. At the cricoid ring, 1 mm of edema results in a 60% reduction in the cross-sectional area of the airway, causing increased airway resistance and increased work of breathing. Laryngomalacia is also common in premature infants and can result in obstruction. Given a large surface area, small body volume, and minimal insulation, neonates are extremely prone to heat loss. Any degree of cold stress is detrimental and increases metabolic demands in the neonate. Infants are unable to shiver effectively, and cold stress causes catecholamine release, which stimulates nonshivering thermogenesis by brown fat. The increased catechols can be detrimental, causing increased peripheral vascular resistance and SVR, increased cardiac stress, and increased oxygen consumption. Anesthesia blunts thermoregulatory sensitivity8 and interferes with nonshivering thermogenesis and brown fat metabolism.19 Anesthesia also increases heat loss by inducing cutaneous vasodilation. At all times, including during transport and in the operating room, the infant must be subjected to a neutral thermal environment. An environment that is too warm can be equally detrimental. Core temperature must be carefully monitored at all times in the operating room. Whole-body glucose demand corrected for body mass in neonates can be twice that in adults. Carbohydrate reserves, primarily hepatic glycogen, in a normal newborn are relatively low, and even lower in an infant with low birth weight. Hypoglycemia can readily occur if the infant is deprived of a glucose source. Even transient hypoglycemia has been associated with neurologic injury in neonates.49 In contrast with adults, in whom hyperglycemia seems to potentiate brain damage during global and focal ischemia,71 hyperglycemia in neonates may provide some protection from ischemic damage.18,56 Generally, intravenous glucose (10% dextrose) or intravenous hyperalimentation should be continued intraoperatively, especially in infants with low birth weight without adequate glycogen or fat stores. Nonetheless, hyperglycemia can also be a problem in the perioperative period. Insulin response is deficient in preterm infants, and high catecholamines owing to illness or intraoperative stress can result in hyperglycemia. Careful monitoring of serum glucose is indicated. Hypoxia is a common pathologic condition in neonates because of the incidence of lung disease; congenital heart disease and disordered transition are other frequent causes. Tissue hypoxia, or oxygen delivery to the tissues inadequate to sustain oxidative metabolism, is harmful, with well-known pathologic sequelae. There is at present no good method for reliably detecting tissue hypoxia, particularly in a real-time manner. Oxygen therapy is widely used, particularly during anesthesia, to avoid hypoxia. Until more recently, oxygen has generally been given in excess in the operating room, almost universally resulting in hyperoxemia. Hyperoxemia is always an iatrogenic condition, and there is increasing evidence that it is causally associated with multiple pathologic conditions in neonates, and should be strenuously avoided.58,73 Oxygen is a highly reactive molecule, and free radical formation by oxygen is well known. Reperfusion injury, which refers to damage resulting from free radical formation on reoxygenation after a period of hypoxia, is well established. Retinopathy of prematurity is the most widely known complication of hyperoxia, but there is increasing evidence that oxygen may play a role in chronic lung disease, neurologic impairment, and other pathologic conditions. For retinopathy of prematurity, even a brief hyperoxic period, such as during an anesthetic procedure, may be detrimental.7 There is also evidence that wide fluctuations in oxygenation, as often experienced during anesthesia, may be damaging. Monitoring of oxygenation during operative procedures is problematic. Continuous pulse oximetry is the standard of care and has contributed to the safety of anesthesia by avoidance of hypoxia. The pulse oximeter is not as helpful for detection of hyperoxia. At hemoglobin saturations greater than 95%, most infants are hyperoxic; this is altered further by varying amounts of fetal hemoglobin. Maintenance of saturations greater than 95% is almost universal in most operating rooms, but is probably not the best course for neonates. It has been suggested that for ongoing management of neonates in the NICU, oxygen saturation targets of 90% to 95% should be used.24,76a Similarly, it has become common practice to use blended oxygen levels of 30% to 40% during initial resuscitation. It seems reasonable that Fio2 during operative procedures should be kept as low as possible, as long as oxygen saturation is 90% or greater, at least as an initial target. Many neonates can be managed intraoperatively on room air. Further consideration probably should be given to the universal practice of using 100% oxygen during induction and emergence from anesthesia, although this is more problematic because hypoxia also should be avoided, and oxygen desaturation during induction and intubation is frequent. This possibility further underscores the need for experienced personnel during induction and intubation of neonates. The time of the last oral feed should be ascertained, especially in elective surgery in healthier infants. Attenuation of airway reflexes with induction of anesthesia places the infant at risk of aspiration of acidic gastric contents, which can lead to serious inflammatory pneumonitis. Nil per os (NPO; nothing by mouth) guidelines are summarized in Table 41-223,69; these guidelines are more liberal for newborns than adults. Studies suggest that a safe NPO time after a formula feed might be 4 hours.13 This shorter length of time partially reflects more rapid gastric emptying, although in an infant, dehydration occurs much more quickly than in an adult, and this is a concern in an infant without an intravenous line. As in adults, gastric emptying is delayed with stress, anxiety, and illness. Infants with diseases such as pyloric stenosis, duodenal atresia, malrotation, or other obstructive lesions are NPO before surgery, and they need an intravenous line to maintain appropriate hydration. Despite the NPO status, there usually are significant gastric contents, and rapid-sequence or awake intubation is required. Emptying the stomach via nasogastric drainage is undependable and usually inadequate. Nasogastric feeds in intubated infants also should be discontinued at an appropriate time because the infant may need to be reintubated during surgery, and a full stomach would increase the risk for aspiration. TABLE 41-2 Physical examination includes vital signs, including temperature. Volume status needs to be carefully assessed because major shifts can occur easily during surgery. Many infants requiring surgery are ill. Infants with congenital anomalies often have multiple defects, which may have an impact on anesthetic care. Often, a group of defects suggests a well-recognized and characterized syndrome. Common syndromes with their anesthetic implications are listed in Table 41-3. (This list is incomplete.) TABLE 41-3 Anesthetic Implications of Neonatal Syndromes
Anesthesia in the Neonate
Complication
Control
Fentanyl
Frequent bradycardia
4
1
Hypotension, poor circulation
4
0
Glycosuria
1
0
Acidosis
2
0
Increased ventilatory requirements
4
1
Intraventricular hemorrhage
2
0
Total complications
17
2
Anesthesia, Neonatal Physiology, and Specific Concerns
Transition Phase and Persistent Pulmonary Hypertension of the Neonate
Respiratory Physiology: Apnea, Central Control of Ventilation, and Respiratory Distress Syndrome
Airway Anatomy
Temperature Regulation
Carbohydrate Metabolism
Oxygen Therapy and Retinopathy of Prematurity and Chronic Lung Disease
Conduct of an Anesthetic
Preoperative Evaluation and Preparation
Intake
Time (h)
Clear liquids
2
Breast milk
4
Formula
6
Syndrome
Clinical Manifestations
Anesthetic Implications
Adrenogenital syndrome
Analbuminemia
Andersen syndrome
Apert syndrome
Chiari malformation
Arthrogryposis multiplex congenita
Beckwith-Wiedemann syndrome
CHARGE association
Cherubism
Cornelia de Lange syndrome
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access