Anesthesia is classically defined as a drug-induced state that includes analgesia, amnesia, and muscle relaxation. The provision of anesthesia to infants undergoing surgical procedures has gone through a remarkable transition coincident with the development of new intravenous agents and more sophisticated monitoring techniques. As recently as 1985, there was considerable debate about whether neonates feel pain, and sophisticated researchers advocated the use of minimal anesthesia in neonates undergoing surgical procedures, citing the dangers of anesthetic administration to this population (25
). Beginning with the landmark paper of Robinson and Gregory (26
), practitioners of neonatal and pediatric anesthesia have proclaimed the importance of providing adequate anesthesia, particularly to ill preterm infants. In modern anesthetic practice, adequate anesthetic depth and control of the neonatal stress response can be achieved without undue acute risk to the infant.
The appropriate anesthetic technique is dictated by the preoperative condition of the patient, planned surgical procedure, and skills of the anesthetist. The encounter between the anesthesiologist and neonate often occurs in the setting of a surgical emergency, and a general anesthetic with control of the airway is most often the technique of choice. General anesthesia is provided using a combination of inhaled and intravenous agents and muscle relaxants.
The surgical stress responses of neonates can be inhibited by potent anesthesia, as demonstrated by randomized trials of halothane anesthesia in term neonates, fentanyl anesthesia in preterm neonates, and sufentanil anesthesia in neonates undergoing cardiac surgery (8
). These results imply that the nociceptive stimuli during surgery are at least partially responsible for the marked stress responses of neonates and are prevented by the provision of adequate anesthesia. In these trials, the reduction in surgical stress responses was associated with significant improvements in clinical outcome, supporting the use of potent anesthetic agents for newborns undergoing surgery. In contrast, a randomized, double-blind study of fentanyl bolus, fentanyl infusion, and fentanyl-midazolam infusion in infants less than 6 months old undergoing cardiac surgery found that these anesthetic regimens did not blunt the metabolic and hormonal responses to surgical stress, yet no adverse postoperative outcomes were observed. These findings indicate that improvements in clinical outcome observed in previous trials may have been related to factors other than reduction in surgical stress (29
The inhaled agents include an inorganic gas (e.g., nitrous oxide) and volatile liquids (e.g., halothane, enflurane, isoflurane, sevoflurane, and desflurane). Delivery of potent inhaled agents by means of the respiratory system offers a reliable route of administration and excretion with the ability to rapidly alter anesthetic concentrations in the central nervous system. Each of the inhaled anesthetic agents has unique effects on the cardiovascular, respiratory, and central nervous systems, which are not exhaustively reviewed here (Table 53.1
). The volatile anesthetics produce dose-dependent decreases in mean arterial blood pressure, particularly in premature infants, due to direct myocardial depression, and decreases in systemic vascular resistance due to exaggerated depression of the baroreceptor reflex (30
). Nitrous oxide produces minimal alterations in myocardial performance or systemic vascular resistance, due in part to direct stimulation of the sympathetic nervous system. However, if combined with a potent volatile agent or opioids, nitrous oxide significantly depresses myocardial contractility (32
All inhaled agents increase the respiratory rate, reduce tidal volume and functional residual capacity, decrease the ventilatory responses to hypoxemia and hypercapnia, and decrease bronchial smooth muscle reactivity. These agents produce a dose-dependent increase in cerebral blood flow, despite simultaneous depression of cerebral metabolic oxygen requirement. At high concentrations, isoflurane and desflurane induce an isoelectric electroencephalographic pattern; this property is not shared by the other inhaled anesthetic agents.
Although halothane is most frequently associated with perioperative hepatic dysfunction, other inhaled agents and intravenous anesthetics may result in hepatic necrosis (33
). The inhaled agents produce dose-related decreases in renal blood flow and urine output due to effects on cardiac output and systemic vascular resistance. Fluoride-induced nephrotoxicity is a potential complication of prolonged exposure to the fluorinated hydrocarbons, although it is of practical concern only during prolonged administration of enflurane and sevoflurane (34
Two newer inhaled anesthetics, sevoflurane and desflurane, are gaining popularity due to their low lipid solubility. This property
allows for rapid induction of anesthesia as well as a short recovery time (35
). Sevoflurane has the advantage of providing a smooth, less irritating induction of anesthesia that rivals that of halothane, with less risk of hepatitis and fewer hemodynamic effects (36
). The drawbacks are the biotransformation of sevoflurane into potentially toxic compound A (2-fluoromethoxy-1,1,3, 3,3-pentafluoro-1-propene) and the accumulation of fluoride ions (37
). As in adults, inhaled sevoflurane has been shown to prolong the QTc interval in infants, an effect that lasts at least 60 minutes into the postoperative period (38
TABLE 53.1 Systemic Effects of Inhaled Anesthetics
Systemic Vascular Resistance
Cerebral Blood Flow
a In doses <1.0 minimum alveolar concentration.
++, greatly increased; +, moderately increased; +/- no consistent effect; -, moderately decreased; – -, greatly decreased.
Morphine and the synthetic opioids have been a consistent adjunct to the volatile agents throughout the history of anesthesia. Highdose opioids have become the preferred anesthetic technique for cardiac surgical procedures in adults and children. The virtues of opioids include minimal effects on myocardial performance, ablation of pulmonary vascular responses to nociceptive stimuli, and preservation of hypoxic pulmonary vasoconstriction (39
Because of their wide margin of safety in ill infants with congenital heart disease, opioid anesthesia is often the anesthesiologist’s choice in preterm infants with cardiopulmonary instability undergoing surgical stress. Fentanyl, sufentanil, and remifentanil are the most popular agents due to their negligible effect on cardiovascular function, but if combined with other anesthetic agents, these opioids may be associated with significant hemodynamic instability. Morphine anesthesia may increase plasma histamine concentrations and decrease vascular resistance, and it is not recommended as a primary anesthetic for ill neonates.
The elimination half-lives of most opioids are variable but significantly prolonged in the neonate (Table 53.2
) and may be further prolonged by any compromise of hepatic blood flow. The exception to this rule is the synthetic opioid remifentanil; a recent study in children showed that clearance was as much as twice as rapid in infants from birth to 2 years than in older children, and half-life was similar in all age groups studied (41
). Premature infants have a more prolonged morphine clearance than term newborns that shortens with postconceptual age (42
). Prolonged postoperative respiratory depression may occur if these important pharmacokinetic variables are ignored in the perioperative period.
The prescription of opiates for their sedative properties has been extended from the operating room to intubated babies in the NICU. Whether infants should be routinely sedated with narcotics while intubated and mechanically ventilated has been investigated in a meta-analysis of 1,505 infants. Although pain scores were significantly improved with opiate use for this purpose, duration of mechanical ventilation, mortality, and short-term and long-term developmental outcomes were not improved. They concluded that opiates should not be used routinely for mechanical ventilation in neonates but selectively when supported by clinical judgment (43
). This clinical judgment is assisted by frequent, regular pain scoring of neonates, which is the standard of practice in modern NICUs. More information about the use of opiates for pain management can be found in the “Analgesia” section below.
TABLE 53.2 Elimination Half-lives of Opioids
Opioid Relative Dose
Regional, Neuraxial, and Local Anesthesia
Regional anesthetic techniques have become increasingly popular in the pediatric and neonatal populations for a number of reasons. General anesthesia may be associated with an increased incidence of postoperative apnea in the preterm infant (44
). This may be a particularly difficult issue in the day-surgery setting, where former preterm neonates commonly present for minor surgical procedures (e.g., circumcision, herniorrhaphy), and in this population, regional or local anesthetic techniques may be particularly advantageous. The use of spinal anesthesia for herniorrhaphy decreased the need for postoperative mechanical ventilation in former preterm infants matched for gestational age at birth and incidence of bronchopulmonary dysplasia (45
). Epidural anesthesia has been shown to reduce the need for postoperative ventilation after both esophageal atresia repair and Nissen fundoplication (46
). Beyond the opioid and mechanical ventilation-sparing effects of regional anesthesia, it may help reduce the use of general anesthetics that have been associated with neurodevelopmental injury in animal models (48
Spinal anesthesia consists of injection of an anesthetic agent into the subarachnoid space. The technique is easy to perform and safe, especially under ultrasound guidance. The most frequent local anesthetic agents are racemic bupivacaine, levobupivacaine, and ropivacaine. Side effects of spinal anesthesia, such as dural puncture headaches and hemodynamic compromise, are common in adults but surprisingly uncommon in infants or children (49
Epidural anesthesia consists of a single injection or repeated injections, through an epidural catheter, of an anesthetic agent into the potential space between the dura mater and ligamentum flavum. The advantage epidural has over spinal anesthesia is the potential for long-term, continuous, or intermittent administration of anesthetics. The most commonly used long-term anesthetic agents are bupivacaine and chloroprocaine, the latter of which has a shorter half-life (51
). Although the epidural space can be approached at any level, for most infants, a lumbar or caudal epidural blockade is used. Epidural catheter placement can be challenging in neonates, but the improvement in ultrasound guidance has helped facilitate easier and safer placement (52
). Though rare, case reports of spinal hematomas serve as a reminder that complications of epidural analgesia can be serious (53
). Caudally inserted epidural catheters can also be advanced to the thoracic region and dilute local anesthetic solutions infused for thoracic-level anesthesia in infants, but radiographic confirmation of tip placement prior to infusion has been shown to be important for safety (54
). A thoracic epidural or surgically placed paravertebral catheter with local anesthetic infusion was found to reduce postoperative ventilator requirement and time to first stool and full enteral feeds, when compared with opiate analgesia in a population of infants undergoing thoracotomy for resection of congenital lung malformations (55
). Caudal epidural blockade with bupivacaine is used most frequently for postoperative pain relief after lower abdominal and lower extremity procedures. Compared with older children and adults, infants and toddlers require higher doses of local anesthetic and demonstrate a shorter duration of effect. Combining local anesthetics with an epidural opioid (fentanyl, hydromorphone), clonidine, or ketamine prolongs the duration of analgesia (56
). However, several case reports have suggested that epidural clonidine may contribute to postoperative apnea in the former preterm infant (57
). Caudal anesthesia has been sufficient as the sole anesthetic technique for lower abdominal procedures but is often used in combination with general anesthesia in infants undergoing abdominal procedures. Rarely, complications result from improper placement of the needle and injection of the anesthetic agent into a vein, the dura, the subarachnoid space, or the sacral marrow.
Local anesthetics may be used to block peripheral nerves in infants undergoing limited surgical procedures (e.g., orchiopexy, herniorrhaphy, circumcision). These techniques are simple to perform, have limited complications, and significantly decrease the need for
postoperative analgesia. Catheters may be left behind to provide continuous infusion of local anesthetic in hopes of extending the duration of the nerve block and provide postoperative analgesia, though care must be taken to not exceed maximum recommended dosing (59
Local anesthetic toxicity is manifested by effects on the cardiovascular system (e.g., myocardial depression, arrhythmias) and central nervous system (e.g., delirium, seizures). In premature infants, the subtle behavioral changes that precede cardiovascular collapse and generalized seizures may be difficult to recognize. The reduced protein binding and prolonged elimination of local anesthetics in this population make the neonate susceptible to toxic effects at lower doses, decreasing the therapeutic index. Careful attention to total administered dose (particularly with field blocks) and monitoring of cardiovascular parameters during the administration of any local anesthetic are essential.
A eutectic mixture of 2.5% lidocaine and 2.5% prilocaine, the topical anesthetic known as EMLA, has shown efficacy in neonatal circumcisions, but two randomized controlled trials have shown that dorsal penile nerve block is more effective than EMLA (60
). EMLA is not effective in reducing pain with heel sticks, perhaps owing to the incidental vasoconstriction associated with EMLA which may result in the need for more vigorous squeezing to obtain a blood sample (62
). Oral sucrose has been shown to be more effective than EMLA as analgesia for neonatal venipuncture in two randomized controlled trials (63
), and a recent trial found that the addition of EMLA to an oral sucrose regimen increased effectiveness of pain control with venipuncture in preterm infants (65
). Although methemoglobinemia is a potential side effect of EMLA, it appears to be safe even in preterm infants.