A. Preoperative Interview
Depending on age, past experiences, and maturity, children present with varying degrees of fright (even terror) when faced with the prospect of surgery or other procedures requiring anesthesia. Unlike adults, who are usually most concerned about the possibility of injury or death, children, when they verbalize their concerns, worry about pain and separation from their parents. Presurgical preparation programs—such as age-appropriate brochures and videos, or tours—can help prepare both children and parents. When time permits, one can demystify the process of anesthesia and surgery by explaining in age-appropriate terms what lies ahead. For example, the anesthesiologist might bring an anesthesia mask for the child to play with during the interview and describe it as like something the astronauts use. Alternatively, in some centers, someone the child trusts (eg, a parent, nurse, another physician) may be allowed to be in attendance during preanesthetic preparations and induction of anesthesia. This can have a particularly calming influence on children undergoing repeated procedures (eg, administration of intrathecal chemotherapy). Unfortunately, outpatient and “same day admit” surgery, coupled with a busy operating room schedule, often make it nearly impossible to adequately reassure parents and patients. Thus, premedication (discussed below) often can be helpful. Some pediatric hospitals have induction rooms adjacent to their operating rooms to permit parental attendance and a quieter, less startling environment for anesthetic inductions.
B. Recent Upper Respiratory Tract Infection
Children frequently present for surgery with signs and symptoms—a runny nose with fever, cough, or sore throat—of a viral upper respiratory tract infection (URI). Attempts should be made to differentiate between an infectious cause of rhinorrhea and an allergic or vasomotor cause. A viral infection within 2 to 4 weeks before general anesthesia and endotracheal intubation appears to place the child at an increased risk for perioperative pulmonary complications, such as wheezing (10-fold), laryngospasm (5-fold), hypoxemia, and atelectasis. This is particularly likely if the child has a severe cough, high fever, or a family history of reactive airway disease. On the other hand, children can have mild URIs on a nearly monthly basis and it can become nearly impossible to schedule them for anesthesia at a time when they neither currently have, nor are recovering from, a URI. The decision to anesthetize children with URIs remains controversial and should be based on the severity of URI symptoms, the urgency of the surgery, and the presence of other coexisting illnesses, When anesthesia will be provided to a child with a URI, one may consider premedicating with an anticholinergic or a β2-agonist (eg, albuterol), avoiding intubation (if feasible), and humidifying inspired gases. A longer-than-usual stay in the postanesthesia recovery area may be required.
Few, if any, preoperative laboratory tests are cost effective. Some pediatric centers require no preoperative laboratory tests in healthy children undergoing minor procedures. Obviously, this places responsibility on the anesthesiologist, surgeon, and pediatrician to correctly identify those patients who should have preoperative testing for specific surgical procedures.
Most asymptomatic patients with cardiac murmurs do not have significant cardiac pathology. Innocent murmurs may occur in more than 30% of normal children. These are typically soft, short systolic ejection murmurs that are best heard along the left upper or left lower sternal border and that do not radiate. Innocent murmurs at the left upper sternal border typically are due to flow across the pulmonic valve (pulmonic ejection) whereas those at the lower left border typically are due to flow from the left ventricle to the aorta (Still’s vibratory murmur). The pediatrician should carefully evaluate patients with a newly diagnosed murmur, particularly in infancy. Consultation with a pediatric cardiologist, echocardiography, or both, should be obtained if the patient is symptomatic (eg, poor feeding, failure to thrive, or easy fatigability). Murmurs that are loud, “harsh,” holosystolic, diastolic, or that radiate widely—or pulses that are either bounding or markedly diminished—require further evaluation and diagnosis.
Because children are more prone to dehydration than adults, their preoperative fluid restriction has always been more lenient. Several studies, however, have documented low gastric pH (<2.5) and relatively high residual volumes in pediatric patients scheduled for surgery, suggesting that children may be at a greater risk for aspiration than was previously thought. The incidence of aspiration is reported to be approximately 1:1000. There is no evidence that prolonged fasting decreases the risk of aspiration. In fact, several studies have demonstrated lower residual volumes and higher gastric pH in pediatric patients who received clear fluids a few hours before induction (see Chapter 53). The guideline on preoperative fasting produced by the American Society of Anesthesiologists specifies that infants may be fed breast milk up to 4 h before induction, and formula or liquids and a “light” meal may be given up to 6 h before induction. Clear fluids are offered until 2 h before induction. These recommendations are for healthy neonates, infants, and children without risk factors for decreased gastric emptying or aspiration. In any case, there is almost no clinical evidence for the recommendations.
There is great variation in the premedication of pediatric patients. Sedative premedication is generally omitted for neonates and sick infants. Children who appear likely to exhibit uncontrollable separation anxiety may be given a sedative, such as midazolam (0.3–0.5 mg/kg, 15 mg maximum). The oral route is generally preferred because it is less traumatic than intramuscular injection, but it requires 20 to 45 min for effect. Smaller doses of midazolam have been used in combination with oral ketamine (4–6 mg/kg) for inpatients. For uncooperative patients, intramuscular midazolam (0.1–0.15 mg/kg, 10 mg maximum) or ketamine (2–3 mg/kg) with atropine (0.02 mg/kg) may be helpful. Rectal midazolam (0.5–1 mg/kg, 20 mg maximum) or rectal methohexital (25–30 mg/kg of 10% solution) may also be administered in such cases while the child is in the parent’s arms. Some clinicians administer dexmedetomidine (1–2 mcg/kg) or midazolam premedication intranasally. Fentanyl can also be administered as a lollipop (Actiq, 5–15 mcg/kg); however, fentanyl levels continue to rise intraoperatively and can contribute to postoperative analgesia.
In the past, anesthesiologists routinely premedicated young children with anticholinergic drugs to reduce the likelihood of bradycardia. Atropine reduces the incidence of hypotension during induction in neonates and in infants younger than 3 months. Atropine can also prevent accumulation of secretions that can block small airways and endotracheal tubes. Secretions can be particularly troublesome for children with URIs or those who have been given ketamine. Atropine may be administered orally (0.05 mg/kg), intramuscularly, or occasionally rectally. In current practice, most prefer to administer atropine intravenously during induction.
Monitoring requirements for infants and children are generally similar to those for adults with some minor modifications. Alarm limits should be appropriately adjusted. Smaller electrocardiographic electrode pads may be necessary so that they do not encroach on sterile surgical areas. Blood pressure cuffs must be properly fitted. Noninvasive blood pressure monitors have proved to be reliable in infants and children. A precordial or esophageal stethoscope provides an inexpensive means of monitoring heart rate, quality of heart sounds, and airway patency. Finally, monitors may sometimes need to be first attached (or reattached) following induction of anesthesia in less cooperative patients.
Pulse oximetry and capnography assume an even more important role in infants and small children because hypoxia from inadequate ventilation remains a common cause of perioperative morbidity and mortality. In neonates, the pulse oximeter probe should preferably be placed on the right hand or earlobe to measure preductal oxygen saturation. As in adult patients, end-tidal CO2 analysis allows assessment of adequacy of ventilation, changes in cardiac output, confirmation of endotracheal tube placement, and early warning of malignant hyperthermia. Flow-through (mainstream) analyzers are usually less accurate in patients weighing less than 10 kg. Even with aspiration (sidestream) capnographs, the inspired (baseline) CO2 can appear falsely elevated and the expired (peak) CO2 can be falsely low. The degree of error can be minimized by placing the sampling site as close as possible to the distal tip of the endotracheal tube, reducing the length of sampling line, and lowering gas-sampling flow rates (100–150 mL/min). The weight of some flow-through sensors may lead to kinking of the warmed endotracheal tube.
Temperature must be closely monitored in pediatric patients because of their greater risk for malignant hyperthermia and greater susceptibility for intraoperative hypothermia or hyperthermia. The risk of hypothermia can be reduced by maintaining a warm operating room environment (26°C or warmer), by warming and humidifying inspired gases, by using a warming blanket and warming lights, and by warming all intravenous and irrigation fluids. These concerns, while important in all patients, are critically important in newborns. Care must be taken to prevent accidental burns and hyperthermia from overzealous warming efforts.
Invasive monitors (eg, arterial cannulation, central venous catheterization) demand expertise and judgment. Air bubbles should be removed from pressure tubing and small volume flushes should be used to avoid air embolism, unintended heparinization, or fluid overload. The right radial artery is often chosen for cannulation in the neonate because its preductal location mirrors the oxygen content of the carotid and retinal arteries. A femoral artery catheter may be a suitable alternative in very small neonates. Left radial or right or left dorsalis pedis arteries are other alternatives. Critically ill neonates may retain an umbilical artery catheter. Internal jugular and subclavian approaches are often used for central lines. Ultrasonography should be used during placement of internal jugular catheters and provides useful information for arterial cannulation as well. Urinary output is an important (but neither sensitive nor specific) indicator of the adequacy of intravascular volume and cardiac output. Noninvasive monitors of stroke volume have only recently been tested in infants and young children.
Premature or small-for-gestational-age neonates, and neonates who have received total parenteral nutrition or whose mothers are diabetic, are prone to hypoglycemia. These infants should have frequent blood glucose measurements: levels below 30 mg/dL in the neonate, below 40 mg/dL in infants, and below 60 mg/dL in children (and below 80 mg/dL in adults) indicate hypoglycemia requiring immediate treatment. Blood sampling for arterial blood gases, hemoglobin, potassium, and ionized calcium concentration can be invaluable in critically ill patients, particularly in those undergoing major surgery or who may be receiving transfusions.
General anesthesia is usually induced by an intravenous or inhalational technique. Induction with intramuscular ketamine (5–10 mg/kg) is reserved for specific situations, such as those involving combative, particularly mentally challenged, or autistic patients. Intravenous induction is usually preferred when the patient comes to the operating room with a functional intravenous catheter or will allow awake venous cannulation. Prior application of EMLA (eutectic mixture of local anesthetic) cream (see Chapter 16) may render intravenous cannulation less painful for the patient, and less stressful for the parent and anesthesiologist. However, EMLA cream is neither a perfect nor a complete solution. Some children become anxious at the sight of a needle, particularly those who have had multiple needle punctures in the past, with or without EMLA. Furthermore, it can be difficult to anticipate in which extremity intravenous cannulation will prove to be successful. Finally, to be effective, EMLA cream must remain in contact with the skin for at least 30 to 60 min. Awake or sedated-awake intubation with topical anesthesia should be considered for emergency procedures in neonates and small infants when they are critically ill or a potential difficult airway is present.
The same induction sequence can be used as in adults: propofol (2–3 mg/kg) followed by a nondepolarizing muscle relaxant (eg, rocuronium, cisatracurium, atracurium), or succinylcholine. We recommend that atropine be given routinely prior to succinylcholine. The advantages of an intravenous technique include availability of intravenous access if emergency drugs need to be administered and rapidity of induction in the child at risk for aspiration. Alternatively (and very commonly in pediatric practice), intubation can be accomplished after the combination of propofol, lidocaine, and an opiate, with or without an inhaled agent, avoiding the need for a paralytic agent. Finally, paralytic agents are not needed for placement of LMAs, which are commonly used in pediatric anesthesia.
Many children do not arrive in the operating room with an intravenous line in place and nearly all dread the prospect of being stuck with a needle. Fortunately, sevoflurane can render small children unconscious within minutes. We find this easier in children who have been sedated (most often with oral midazolam) prior to entering the operating room and who are sleepy enough to be anesthetized without ever knowing what has happened (steal induction). One can also insufflate the anesthetic gases over the face, place a drop of food flavoring on the inside of the mask (eg, oil of orange), and allow the child to sit during the early stages of induction. Specially contoured masks minimize dead space (see Figure 19–11).
There are many differences between adult and pediatric anatomy that influence mask ventilation and intubation. Equipment appropriate for age and size should be selected (Table 42–6). Neonates and most young infants are obligate nasal breathers and obstruct easily. Oral airways will help displace an oversized tongue; nasal airways, so useful in adults, can traumatize small nares or prominent adenoids in small children. Compression of submandibular soft tissues should be avoided during mask ventilation to prevent upper airway obstruction.
TABLE 42–6Sizing of airway equipment in children. ||Download (.pdf) TABLE 42–6 Sizing of airway equipment in children.
| ||Premature ||Neonate ||Infant ||Toddler ||Small Child ||Large Child |
|Age ||0–1 month ||0–1 month ||1–12 months ||1–3 years ||3–8 years ||8–12 years |
|Weight (kg) ||0.5–3 ||3–5 ||4–10 ||8–16 ||14–30 ||25–50 |
|Tracheal (ET)1 tube (mm i.d.) ||2.5–3 ||3–3.5 ||3.5–4 ||4–4.5 ||4.5–5.5 ||5.5–6 (cuffed) |
|ET depth (cm at lips) ||6–9 ||9–10 ||10–12 ||12–14 ||14–16 ||16–18 |
|Suction catheter (F) ||6 ||6 ||8 ||8 ||10 ||12 |
|Laryngoscope blade ||00 ||0 ||1 ||1.5 ||2 ||3 |
|Mask size ||00 ||0 ||0 ||1 ||2 ||3 |
|Oral airway ||000–00 ||00 ||0 (40 mm) ||1 (50 mm) ||2 (70 mm) ||3 (80 mm) |
|Laryngeal mask airway (LMA#) ||— ||1 ||1 ||2 ||2.5 ||3 |
Typically, the child can be coaxed into breathing an odorless mixture of nitrous oxide (70%) and oxygen (30%). Sevoflurane (or halothane) can be added to the gas mixture in 0.5% increments every few breaths. As previously discussed, we favor sevoflurane in most situations. Desflurane and isoflurane are avoided for inhalation induction because they are pungent and associated with more coughing, breath-holding, and laryngospasm. We use a single (sometimes two) breath induction technique with sevoflurane (7–8% sevoflurane in 60% nitrous oxide) to speed the induction in cooperative patients. After an adequate depth of anesthesia has been achieved, an intravenous line can be started and propofol and an opioid (or a muscle relaxant) administered to facilitate intubation. Patients typically pass through an excitement stage during which any stimulation can induce laryngospasm. Breath-holding must be distinguished from laryngospasm. Steady application of 10 cm of positive end-expiratory pressure will usually overcome laryngospasm.
Alternatively, the anesthesiologist can deepen the level of anesthesia, by increasing the concentration of volatile anesthetic, and place an LMA or intubate the patient under “deep” sevoflurane anesthesia. Because of the greater anesthetic depth required for tracheal intubation, the risk of cardiac depression, bradycardia, or laryngospasm occurring without intravenous access detracts from this latter technique. Intramuscular succinylcholine (4 mg/kg, not to exceed 150 mg) and atropine (0.02 mg/kg, not to exceed 0.4 mg) should be available if laryngospasm or bradycardia occurs before an intravenous line is established.
Positive-pressure ventilation during mask induction and prior to intubation sometimes causes gastric distention, with impairment of lung expansion. Suctioning with an orogastric or nasogastric tube will decompress the stomach, but it must be done without traumatizing fragile mucous membranes.
Intravenous cannulation in infants can be a vexing ordeal. This is particularly true for infants who have spent weeks in a neonatal intensive care unit and have few intact veins. Even healthy 1-year-old children can prove a challenge because of extensive subcutaneous fat. Venous cannulation usually becomes easier after 2 years of age. The saphenous vein has a consistent location at the ankle and an experienced practitioner can usually cannulate it even if it is not visible or palpable. Transillumination of the hands or ultrasonography will often reveal previously hidden cannulation sites. Twenty-four-gauge over-the-needle catheters are adequate in neonates and infants when blood transfusions are not anticipated. All air bubbles should be removed from the intravenous line to reduce the risk of paradoxical air embolism from occult patent foramen ovale. In emergency situations where intravenous access is impossible, fluids can be effectively infused through an 18-gauge needle inserted into the medullary sinusoids within the tibial bone. This intraosseous infusion can be used for all medications normally given intravenously, with almost as rapid results (see Chapter 55), and is considered part of the standard trauma resuscitation, advanced cardiac life support (ACLS), and pediatric advanced life support (PALS) protocols when intravenous access cannot be obtained.
One hundred percent oxygen should be administered prior to intubation to increase patient safety during the obligatory period of apnea prior to and during intubation. The choice of muscle relaxant has been discussed earlier in this chapter. For awake intubations in neonates or infants, adequate preoxygenation and continued oxygen insufflation during laryngoscopy may help prevent hypoxemia.
The infant’s prominent occiput tends to place the head in a flexed position prior to intubation. This is easily corrected by slightly elevating the shoulders on towels and placing the head on a doughnut-shaped pillow. In older children, prominent tonsillar tissue can obstruct visualization of the larynx. Straight laryngoscope blades aid intubation of the anterior larynx in neonates, infants, and young children (Table 42–6). Endotracheal tubes that pass through the glottis may still impinge upon the cricoid cartilage, which is the narrowest point of the airway in children younger than 5 years of age. Mucosal trauma from trying to force a tube through the cricoid cartilage can cause postoperative edema, stridor, croup, and airway obstruction.
The appropriate diameter inside the endotracheal tube can be estimated by a formula based on age:
4 + Age/4 = Tube diameter (in mm)
For example, a 4-year-old child would be predicted to require a 5-mm uncuffed tube. This formula provides only a rough guideline, however. Exceptions include premature neonates (2.5–3 mm tube) and full-term neonates (3–3.5 mm tube). Alternatively, the practitioner can remember that a newborn takes a 2.5- or 3-mm tube, and a 5-year-old takes a 5-mm tube. It should not be that difficult to identify which of the three sizes of tube between 3 and 5 mm is required in small children. In larger children, small (5–6 mm) cuffed tubes can be used either with or without the cuff inflated to minimize the need for precise sizing. Endotracheal tubes 0.5 mm larger and smaller than predicted should be readily available in or on the anesthetic cart. In the past, uncuffed endotracheal tubes were selected for children aged 5 years or younger in the hope of decreasing the risk of postintubation croup. Currently, many anesthesiologists no longer use size 4.0 or larger uncuffed tubes. The leak test will minimize the likelihood that an excessively large tube has been inserted. Correct tube size and appropriate cuff inflation is confirmed by easy passage into the larynx and the development of a gas leak at 15 to 25 cm H2O pressure. No leak indicates an oversized tube or overinflated cuff that should be replaced or deflated to prevent postoperative edema, whereas an excessive leak may preclude adequate ventilation and contaminate the operating room with anesthetic gases. As previously noted, many clinicians use a down-sized cuffed tube in younger patients at high risk for aspiration; minimal inflation of the cuff can stop any air leak. There is also a formula to estimate endotracheal length:
12 + Age/2 = Length of tube (in cm)
Again, this formula provides only a guideline, and the result must be confirmed by auscultation and clinical judgment. To avoid endobronchial intubation, the tip of the endotracheal tube should pass only 1 to 2 cm beyond an infant’s glottis. Alternatively, one can intentionally advance the tip of the endotracheal tube into the right mainstem bronchus and then withdraw it until breath sounds are equal over both lung fields.
Ventilation is almost always controlled during anesthesia of neonates and infants when using a conventional semiclosed circle system. During spontaneous ventilation, even the low resistance of a circle system can become a significant obstacle for a sick neonate to overcome. Unidirectional valves, breathing tubes, and carbon dioxide absorbers account for most of this resistance. For patients weighing less than 10 kg, some anesthesiologists prefer the Mapleson D circuit or the Bain system because of their low resistance and light weight (see Chapter 3). Nonetheless, because breathing-circuit resistance is easily overcome by positive-pressure ventilation, the circle system can be safely used in patients of all ages if ventilation is controlled. Monitoring of airway pressure may provide early evidence of obstruction from a kinked endotracheal tube or accidental advancement of the tube into a mainstem bronchus.
Many anesthesia ventilators on older machines are designed for adult patients and cannot reliably provide the reduced tidal volumes and rapid rates required by neonates and infants. Unintentional delivery of large tidal volumes to a small child can generate excessive peak airway pressures and cause barotrauma. Pressure control ventilation, which is found on nearly all newer anesthesia ventilators, should be used for neonates, infants, and toddlers. Small tidal volumes can also be manually delivered with greater ease with a 1-L breathing bag than with a 3-L adult bag. For children less than 10 kg, adequate tidal volumes are achieved with peak inspiratory pressures of 15 to 18 cm H2O. For larger children the volume control ventilation may be used and tidal volumes may be set at 6 to 8 mL/kg. Many spirometers are less accurate at lower tidal volumes. In addition, the gas lost in long, compliant adult breathing circuits becomes large relative to a child’s small tidal volume. For this reason, pediatric breathing circuits are usually shorter, lighter, and stiffer (less compliant). Nevertheless, one should recall that the additional dead space contributed by the tube and circle system consists only of the volume of the distal limb of the Y-connector and that portion of the endotracheal tube that extends beyond (proximal to) the airway. In other words, the dead space is unchanged by switching from adult to pediatric breathing circuits. Condenser humidifiers or heat and moisture exchangers (HMEs) can add considerable dead space; depending on the size of the patient, they either should not be used or an appropriately sized, pediatric HME should be employed.
Anesthesia can be maintained in pediatric patients with the same agents as in adults. Some clinicians switch to isoflurane following a sevoflurane induction in the hope of reducing the likelihood of emergence agitation or postoperative delirium (see earlier discussion). Administration of an opioid (eg, fentanyl, 1–1.5 mcg/kg) or dexmedetomidine (0.5 mcg/kg, given slowly with heart rate monitoring) 15 to 20 min before the end of the procedure can reduce the incidence of emergence delirium and agitation if the surgical procedure is likely to produce postoperative pain. Although the MAC is greater in children than in adults (see Table 42–4), neonates may be particularly susceptible to the cardiac-depressing effects of general anesthetics and may not tolerate the concentrations of volatile agents required when the volatile agent alone is used to maintain good surgical operating conditions.
Perioperative Fluid Requirements
Meticulous attention to fluid intake and loss is required in younger pediatric patients because these patients have limited margins for error. A programmable infusion pump or a buret with a microdrip chamber is useful for accurate measurements. Drugs can be flushed through low dead-space tubing to minimize unnecessary fluid administration. Fluid overload is diagnosed by prominent veins, flushed skin, increased blood pressure, decreased serum sodium, and a loss of the folds in the upper eyelids.
Fluid therapy can be divided into maintenance, deficit, and replacement requirements.
A. Maintenance Fluid Requirements
Maintenance requirements for pediatric patients can be determined by the “4:2:1 rule”: 4 mL/kg/h for the first 10 kg of weight, 2 mL/kg/h for the second 10 kg, and 1 mL/kg/h for each remaining kilogram. The choice of maintenance fluid remains controversial. A solution such as D5½ NS with 20 mEq/L of potassium chloride provides adequate dextrose and electrolytes at these maintenance infusion rates. D5¼ NS may be a better choice in neonates because of their limited ability to handle sodium loads. Children up to the age of 8 years require 6 mg/kg/min of glucose to maintain euglycemia (40–125 mg/dL); premature neonates require 6–8 mg/kg/min. Euglycemia is normally well maintained in older children and adults by hepatic glycogenolysis and gluconeogenesis despite administration of glucose-free solutions. Both hypoglycemia and hyperglycemia should be avoided; however, the amount of hepatic glucose production is widely variable during major surgery and critical illness. Thus glucose infusion rates during longer surgeries, particularly in neonates and infants, should be adjusted based on blood glucose measurements.
In addition to a maintenance infusion, any preoperative fluid deficits must be replaced. For example, if a 5-kg infant has not received oral or intravenous fluids for 4 h prior to surgery, a deficit of 80 mL has accrued (5 kg × 4 mL/kg/h × 4 h). In contrast to adults, infants respond to dehydration with decreased blood pressure and without increased heart rate. Preoperative fluid deficits are often administered with hourly maintenance requirements in aliquots of 50% in the first hour and 25% in the second and third hours. In the example above, a total of 60 mL would be given in the first hour (80/2 + 20) and 40 mL in the second and third hours (80/4 + 20). Bolus administration of dextrose-containing solutions should be avoided to prevent hyperglycemia. Preoperative fluid deficits are usually replaced with a balanced salt solution (eg, lactated Ringer’s injection) or ½NS. Glucose is omitted to prevent hyperglycemia. Compared with lactated Ringer’s injection, normal saline has the disadvantage of promoting hyperchloremic acidosis.
C. Replacement Requirements
Replacement can be subdivided into blood loss and third-space loss.
The blood volume of premature neonates (100 mL/kg), full-term neonates (85–90 mL/kg), and infants (80 mL/kg) is proportionately larger than that of adults (65–75 mL/kg). An initial hematocrit of 55% in the healthy full-term neonate gradually falls to as low as 30% in the 3-month-old infant before rising to 35% by 6 months. Hemoglobin (Hb) type is also changing during this period: from a 75% concentration of HbF (greater oxygen affinity, reduced PaO2, poor tissue unloading) at birth to almost 100% HbA (reduced oxygen affinity, high PaO2, good tissue unloading) by 6 months.
Blood loss has been typically replaced with non–glucose-containing crystalloid (eg, 3 mL of lactated Ringer’s injection for each milliliter of blood lost) or colloid solutions (eg, 1 mL of 5% albumin for each milliliter of blood lost) until the patient’s hematocrit reaches a predetermined lower limit. In recent years there has been increased emphasis on avoiding excessive fluid administration; thus blood loss is now commonly replaced by either colloid (eg, albumin) or packed red blood cells. In premature and sick neonates, the target hematocrit (for transfusion) may be as great as 40%, whereas in healthy older children a hematocrit of 20% to 26% is generally well tolerated. Because of their small intravascular volume, neonates and infants are at an increased risk for electrolyte disturbances (eg, hyperglycemia, hyperkalemia, and hypocalcemia) that can accompany rapid blood transfusion. Dosing of packed red blood cell transfusions is discussed in Chapter 51. Platelets and fresh frozen plasma, 10 to 15 mL/kg, should be given when blood loss exceeds one to two blood volumes. Recent practice, particularly with blood loss from trauma, favors “earlier” administration of plasma and platelets as part of a massive transfusion protocol. One unit of platelets per 10 kg weight raises the platelet count by about 50,000/μL. The pediatric dose of cryoprecipitate is 1 unit/10 kg weight.
These losses are impossible to measure and must be estimated by the extent of the surgical procedure. In recent years some investigators have questioned the very existence of the third space, and some have asserted that the third space exists as a consequence of excessive fluid administration.
One popular fluid administration guideline is 0 to 2 mL/kg/h for relatively atraumatic surgery (eg, strabismus correction where there should be no third-space loss) and up to 6 to 10 mL/kg/h for traumatic procedures (eg, abdominal abscess). Third-space loss is usually replaced with lactated Ringer’s injection (see Chapter 49). It is safe to say that all issues relating to the third space have never been more controversial.
Regional Anesthesia and Analgesia
The primary uses of regional techniques in pediatric anesthesia have been to supplement and reduce general anesthetic requirements and to provide better postoperative pain relief. Blocks range in complexity from relatively simple peripheral nerve blocks (eg, penile block, ilioinguinal block); to brachial plexus, sciatic nerve, femoral nerve, and transversus abdominis plane (TAP) blocks; to major conduction blocks (eg, spinal or epidural techniques). Regional blocks in children (as in adults) are often facilitated by ultrasound guidance, sometimes with nerve stimulation.
Caudal blocks have proved useful following a variety of surgeries, including circumcision, inguinal herniorrhaphy, hypospadias repair, anal surgery, clubfoot repair, and other subumbilical procedures. Contraindications include infection around the sacral hiatus, coagulopathy, or anatomic abnormalities. The patient is usually lightly anesthetized or sedated and placed in the lateral position.
For pediatric caudal anesthesia, a short-bevel 22-gauge needle can be used. If the loss-of-resistance technique is used, the glass syringe should be filled with saline, not air, because of the latter’s possible association with air embolism. After the characteristic pop that signals penetration of the sacrococcygeal membrane, the needle angle of approach is reduced and the needle is advanced only a few more millimeters to avoid entering the dural sac or the anterior body of the sacrum. Aspiration is used to check for blood or cerebrospinal fluid; local anesthetic can then be slowly injected; failure of a 2-mL test dose of local anesthetic with epinephrine (1:200,000) to produce tachycardia helps exclude intravascular placement.
Many anesthetic agents have been used for caudal anesthesia in pediatric patients, with 0.125% to 0.25% bupivacaine (up to 2.5 mg/kg) or 0.2% ropivacaine being most common. Ropivacaine, 0.2%, can provide analgesia similar to bupivacaine but with less motor blockade. Ropivacaine appears to have less cardiac toxicity than bupivacaine when compared milligram to milligram. Addition of epinephrine to caudal solutions tends to increase the degree of motor block. Clonidine, either by itself or combined with local anesthetics, has also been widely used. Morphine sulfate (25 mcg/kg) or hydromorphone (6 mcg/kg) may be added to the local anesthetic solution to prolong postoperative analgesia for inpatients, but will increase the risk of delayed postoperative respiratory depression. The volume of local anesthetic required depends on the level of blockade desired, ranging from 0.5 mL/kg for a sacral block to 1.25 mL/kg for a midthoracic block. Single-shot injections generally last 4 to 12 h. Placement of 20-gauge caudal catheters with continuous infusion of local anesthetic (eg, 0.125% bupivacaine or 0.1% ropivacaine at 0.2–0.4 mg/kg/h) or an opioid (eg, fentanyl, 2 mcg/mL at 0.6 mcg/kg/h) allows prolonged anesthesia and postoperative analgesia. Complications are rare but include local anesthetic toxicity from increased blood concentrations (eg, seizures, hypotension, arrhythmias), spinal blockade, and respiratory depression. Urinary retention is not a problem following single-dose caudal anesthesia.
Lumbar and thoracic epidural catheters can be placed in anesthetized children using the standard loss-of-resistance technique and either a midline or paramedian approach. In small children, caudal epidural catheters have been passed into a thoracic position with the tip localized radiographically.
Unilateral TAP blocks are commonly used to provide analgesia after hernia repair. Bilateral TAP blocks can be used to provide effective postoperative analgesia after abdominal surgery with a lower midline incision. Rectus sheath blocks can be used for midline incision in the upper abdomen.
Spinal anesthesia has been used in some centers for infraumbilical procedures in neonates and infants. Infants and children typically have minimal hypotension from sympathectomy. Intravenous access can be established (conveniently in the foot) after the spinal anesthetic has been administered. This technique has become more widely used for neonates and infants as the potential risk of neurotoxicity from general anesthesia has received greater attention.
Many children will not tolerate placement of nerve blocks or nerve block catheters while awake; however, most peripheral block techniques can be performed safely in anesthetized children. When the area of operation is the upper extremity, we recommend those brachial plexus procedures that can most readily be performed using ultrasound guidance, specifically axillary, supraclavicular, and infraclavicular blocks. We suggest that interscalene block be performed in anesthetized patients only by those with experience and skill with ultrasound guidance and only for procedures where other block techniques would be inferior (eg, upper shoulder procedures) due to the reported rare occurrence of accidental intramedullary injections when interscalene blocks were performed in anesthetized adults. Single-shot and continuous femoral, adductor canal, and sciatic blocks are easily performed in children using ultrasound guidance. The latter can be performed using either a gluteal or a popliteal approach.
A wide variety of other terminal nerve blocks (eg, digital nerve, median nerve, occipital nerve, etc) are easily performed to reduce postoperative pain in children.
Sedation for Procedures In & Out of the Operating Room
Sedation is often requested for pediatric patients inside and outside the operating room for nonsurgical procedures. Cooperation and motionlessness may be required for imaging studies, bronchoscopy, gastrointestinal endoscopy, cardiac catheterization, dressing changes, and minor procedures (eg, casting and bone marrow aspiration). Requirements vary depending on the patient and the procedure, ranging from anxiolysis (minimal sedation), to conscious sedation (moderate sedation and analgesia), to deep sedation/analgesia, and finally to general anesthesia. Anesthesiologists are held to the same standards whether they provide moderate or deep sedation or they provide general anesthesia. This includes preoperative preparation (eg, fasting), assessment, monitoring, and postoperative care. Airway obstruction and hypoventilation are the most commonly encountered problems associated with moderate or deep sedation. With deep sedation and general anesthesia cardiovascular depression can also be a problem.
Table 42–3 includes doses of sedative-hypnotic drugs. One of the sedatives commonly used by nonanesthesia personnel in the past was chloral hydrate, 25 to 100 mg/kg orally or rectally. It has a slow onset of up to 60 min and a long half-life (8–11 h) that results in prolonged somnolence. Although it generally has little effect on ventilation, it can cause fatal airway obstruction in patients with sleep apnea. Moreover, chloral hydrate is a poor choice given its propensity for producing cardiac arrhythmias when it is used in the larger doses needed for moderate sedation. Midazolam, 0.5 mg/kg orally or 0.1 to 0.15 mg/kg intravenously, is particularly useful because its effects can be readily reversed with flumazenil. Doses should be reduced whenever more than one agent is used because of the potential for synergistic respiratory and cardiovascular depression.
Propofol is by far the most useful sedative-hypnotic drug. Although the drug is not approved for sedation of pediatric ICU patients and is not approved for administration by anyone other than those trained in the administration of general anesthesia, it can be dosed safely for most procedures at infusion rates up to 200 mcg/kg/min. In countries other than the United States, propofol is often administered using the Diprifusor, a computer-controlled infusion pump that maintains a constant target site concentration. Supplemental oxygen and close monitoring of the airway, ventilation, and other vital signs are mandatory (as with other agents). An LMA is usually well tolerated at higher doses.
For imaging studies, intranasal dexmedetomidine has also proven useful, especially with infants who do not have or need intravenous access.
Pediatric patients are particularly vulnerable to two common postanesthetic complications: laryngospasm and postintubation croup. As with adult patients, postoperative pain requires close, careful attention. Pediatric anesthesia practice varies widely, particularly in regard to extubation following a general anesthetic. In some pediatric hospitals, all children who will be extubated after a general anesthetic arrive in the postanesthesia care unit (PACU) with the tube or LMA still in place. They are subsequently extubated by the PACU nurse when defined criteria are reached. In other centers, nearly all children are extubated in the operating room before arriving in the PACU. High quality and safety are reported at centers following either protocol.
Laryngospasm is a forceful, involuntary spasm of the laryngeal musculature caused by stimulation of the superior laryngeal nerve (see Chapter 19). It may occur at induction, emergence, or any time in between without an endotracheal tube. Presumably, it can also occur when a tube is in place, but its occurrence will not be recognized. Laryngospasm is more common in young pediatric patients (almost 1 in 50 anesthetics) than in adults, and is most common in infants 1 to 3 months old. Laryngospasm at the end of a procedure can usually be avoided by extubating the patient either while awake (opening the eyes) or while deeply anesthetized (spontaneously breathing but not swallowing or coughing); both techniques have advocates, and despite strong opinions, evidence is lacking as to which is the better approach. Extubation during the interval between these extremes, however, is generally recognized as more hazardous. Recent URI or exposure to secondhand tobacco smoke predisposes children to laryngospasm on emergence. Treatment of laryngospasm includes gentle positive-pressure ventilation, forward jaw thrust, deepening of the anesthetic with intravenous propofol, intravenous lidocaine (1–1.5 mg/kg), or paralysis with intravenous succinylcholine (0.5–1 mg/kg), or rocuronium (0.4 mg/kg) and controlled ventilation. Intramuscular succinylcholine (4–6 mg/kg) remains an acceptable alternative in patients without intravenous access and in whom conservative measures have failed. Laryngospasm is usually an immediate postoperative event but may occur in the recovery room as the patient wakes up and chokes on pharyngeal secretions. For this reason, recovering pediatric patients should be positioned in the lateral position so that oral secretions pool and drain away from the vocal cords. When the child begins to regain consciousness, having the parents at the bedside may reduce his or her anxiety.
Croup is due to glottic or tracheal edema. Because the narrowest part of the pediatric airway is the cricoid cartilage, this is the most susceptible area. Croup is less common with properly sized endotracheal tubes that are small enough to allow a slight gas leak at 10 to 25 cm H2O. Postintubation croup is associated with early childhood (age 1–4 years), repeated intubation attempts, overly large endotracheal tubes, prolonged surgery, head and neck procedures, and excessive movement of the tube (eg, coughing with the tube in place, moving the patient’s head). Intravenous dexamethasone (0.25–0.5 mg/kg) may prevent formation of edema, and inhalation of nebulized racemic epinephrine (0.25–0.5 mL of a 2.25% solution in 2.5 mL normal saline) is an often-effective treatment. Although postintubation croup occurs later than laryngospasm, it will almost always appear within 3 h after extubation.
C. Postoperative Pain Management
Pain in pediatric patients has received considerable attention in recent years, and over that time the use of regional anesthetic and analgesic techniques (as previously above) has greatly increased. Commonly used parenteral opioids include fentanyl (1–2 mcg/kg), morphine (0.05–0.1 mg/kg), and hydromorphone (15 mcg/kg). A multimodal technique incorporating ketorolac (0.5–0.75 mg/kg) and intravenous dexmedetomidine will reduce opioid requirements. Oral, rectal, or intravenous acetaminophen will also reduce opioid requirements and can be a helpful substitute for ketorolac.
Patient-controlled analgesia (see Chapter 48) can also be successfully used in patients as young as 5 years old, depending on their maturity and on preoperative preparation. Commonly used opioids include morphine and hydromorphone. With a 10-min lockout interval, the recommended interval dose is either morphine, 20 mcg/kg, or hydromorphone, 5 mcg/kg. As with adults, continuous infusions increase the risk of respiratory depression; typical continuous infusion doses are morphine, 0 to 12 mcg/kg/h, or hydromorphone, 0 to 3 mcg/kg/h. The subcutaneous route may be used with morphine. Nurse-controlled and parent-controlled analgesia remain controversial but widely used techniques for pain control in children.
As with adults, epidural infusions for postoperative analgesia often consist of a local anesthetic combined with an opioid. Bupivacaine, 0.1% to 0.125%, or ropivacaine, 0.1% to 0.2%, are often combined with fentanyl, 2 to 2.5 mcg/mL (or equivalent concentrations of morphine or hydromorphone). Recommended infusion rates depend on the size of the patient, the final drug concentration, and the location of the epidural catheter, and range from 0.1 to 0.4 mL/kg/h. Local anesthetic infusions can also be used with continuous nerve block techniques, but this is less common than in adults.