Radiant Warmers

Overhead radiant heating units with servomechanism temperature control were useful to maintain the temperature of small neonates and infants in the past, but these have been replaced with other strategies. Given the risk of skin burn, the servomechanism control sensor must be applied to the warmed skin and should not measure body core temperature.¹ A maximum skin temperature of 98.6° F (37° C) should preclude surface burns, although the risk persists in poorly perfused skin. The radiant warmer may be used during induction and surgical preparation; it may be used again as the drapes are removed. If infrared light bulbs are used, care must be taken to ensure that the heat lamp is an appropriate distance from the child to avoid thermal injury²; note that the red cast from the light may make evaluation of a child’s color difficult.

Warming Blankets

Circulating water mattresses help maintain normothermia in children with body surface areas of 0.5 m² or less (approximately 10 kg).³ For most children, conductive heat loss accounts for less than 5% of total body heat loss; therefore these devices have limited usefulness in maintaining thermal homeostasis and virtually no benefit for children weighing more than 10 kg. To avoid surface burns, the fluid temperature should never exceed 102.2° F (39° C) and must be monitored⁴; several layers of material should be interposed between the child and the warming blanket to avoid direct contact. In many institutions these have been replaced by forced air warming mattresses.

Forced Warm Air Devices

The most useful device for maintaining thermal homeostasis is the warm air mattress that can be wrapped around the head and upper or lower torso.^5–9 These devices rely on the combination of convection with warm air and plastic wrap that also serves to reduce evaporative heat losses. They are the single most effective means for warming children, even when only a portion of the body is covered.^5,10 One report suggests that simple delivery of warm air beneath bed sheets without the use of a plastic blanket can be effective,¹⁰ but significant thermal injuries have been reported without the use of a licensed blanket or when used in children with poor skin perfusion.⁶ Therefore unlicensed use of these devices is not recommended. In extreme cases of hypothermia (cardiac surgery or liver transplant), a blanket can be placed both beneath and on top of the child to speed the warming process. Two concerns have been raised regarding these devices. First, concern has been expressed that these devices may incubate infections and distribute them into the OR. The standard design and design changes in the hardware may, unbeknownst to the operator, increase the infectious risk from some warmers.^7–9 Concern also exists that under laminar airflow conditions for joint replacement, these devices increase the infectious risk.¹¹ Also, the cost of these disposable air mattresses may be a considerable expense.

Passive Heat and Moisturizer Exchangers, “Artificial Nose”

Heat moisture exchangers are effective in preserving body heat, but ineffective to raise the child’s body temperature.^12–14 In infants and children, these devices add humidity to the circuit¹⁵ and may also function as filters for infectious pathogens and reduce the transmission of infectious agents between children.^16,17 These devices may increase the resistance to breathing during spontaneous ventilation and completely obstruct the circuit with prolonged use.^18–20

Fluid and Blood Warmers

The effectiveness of standard fluid warmers depends on the time that the intravenous fluids or blood products are in contact with the warmer. Excessive heating may cause hemolysis of red blood cells.²¹ With slow intravenous fluid therapy, the fluid warmer does not contribute significant heat transfer because heat is lost along the intravenous tubing between the warmer and the child, before the fluid enters the child’s body. Warming intravenous fluid during maintenance therapy is virtually useless for temperature regulation.

Blood warmers vary from simple coiled tubing in a warming bath to the more sophisticated rapid transfusion devices. Rapid transfusion systems are vitally important in any institution that manages trauma patients or performs major surgical procedures. The new generation of warming devices is far superior to waterbath warmers. They use countercurrent heat exchange or microwave technology as a means of rapidly warming fluid along the length of the column of fluid. The choice of device depends on the size of the child, the anticipated rate of blood loss, and cost. The length of the intravenous tubing and the diameter of the intravenous catheter also limit the rate of rapid infusion. An example of these devices with the lowest capacity is the Hot Line (Smiths Medical ASD, Rockland, Mass.), which is extremely effective at rates up to 75 mL/min.²² The manufacturer’s data suggest that approximately 90 mL/min of refrigerated blood products can be warmed to 95° F (35° C) with this device. Another device that has a low volume (6 mL priming volume) and is inserted within the intravenous system just before the intravenous catheter is the Belmont Buddy fluid warmer (Belmont Instrument Corp., Billerica, Mass.). The manufacturer claims a warming capacity of cold fluids to 100.4° F (38° C) at flow rates as low as “keep vein open” (KVO) and as great as 100 mL/min.²³ This device should be inserted close to the intravenous catheter because it may help maintain the child’s temperature by minimizing heat loss along the length of the intravenous tubing. A more complex and larger capacity system is the Level 1 System 1000 (Smiths Medical ASD, Rockland, Mass.), which uses countercurrent warming technology combined with the capacity to infuse fluid, blood, or plasma in bags under pressure. Because this device infuses bags of fluid under pressure, it is important to eliminate all air from the bags to avoid the possibility of air embolization. This system is capable of warming fluids at rates up to approximately 500 mL/min.²⁴ The Level 1 device is capable of warming blood starting at 41° F (5° C) to 42.8° F (6° C) to at least 91.4° F (33° C) at rates as high as 250 mL/min.²⁵ The manufacturer’s data suggest that up to 600 mL/min of red cells diluted to a hematocrit of 60% can be delivered through a 14-gauge intravenous catheter and 750 mL/min through an 8.5F introducer sheath. The addition of an air detection device with automatic flow interruption has improved safety.²⁶ Another high-capacity system is the Belmont FMS warmer (Belmont Instrument Corp, Billerica, Mass.) uses microwave warming and is capable of fluid delivery at rates of 10 to 750 mL/min; this device has two air detectors and an in-line pressure detector. One study compared the flow rates and warming capabilities of the Level 1 system with the Rapid Infusion System (no longer manufactured but equivalent to two Belmont FMS systems); both devices were comparable when used with 18- or 20-gauge intravenous catheters in terms of warming and flow rates. However, for any catheter larger than 18 gauge, the Rapid Infusion System was superior for both flow and fluid warming, particularly for flows greater than 200 mL/min (Figs. 51-1 and 51-2).^27,28 A similar comparison has been made with the Level 1 and the Belmont FMS systems, with a similar difference in terms of better warming capacity during high flows.²⁹

FIGURE 51-1 This figure plots the mean temperature of the fluid at the end of two 2-liter infusions of crystalloid for the Level 1 versus the Rapid Infusion System. Note that both devices have equivalent warming capacities with flow rates 200 mL or less per minute but that there is markedly less warming capacity with the Level 1 system at higher flow rates.

(From Barcelona SL, Vilich F, Coté CJ. A comparison of flow rates and warming capabilities of the Level 1 and Rapid Infusion System with various-size intravenous catheters. Anesth Analg 2003;97:358-63.)

FIGURE 51-2 Percent differences in flow rates for the Rapid Infusion System versus the Level 1 for various catheter sizes. Note that the Rapid Infusion System produced greater incremental changes in flow with progressively larger catheters (16 gauge or larger). The Belmont FMS system should have performance characteristics similar to those of the Rapid Infusion System.

(From Barcelona SL, Vilich F, Coté CJ. A comparison of flow rates and warming capabilities of the Level 1 and Rapid Infusion System with various-size intravenous catheters. Anesth Analg 2003;97:358-63.)

In general, the Level 1 Hotline (Smiths Medical) or the Belmont Buddy appear most suited for low-volume, relatively slow blood loss. The Level 1 system seems most appropriate for children weighing 40 kg or less with expected rapid blood loss and the Belmont FMS for larger children with expected massive and rapid blood loss. The advantages and disadvantages of these warming devices are compared in Table 51-1.³⁰

Heated Humidifiers

Airstream warming and humidification devices are useful for maintaining body temperature.^31–33 If a humidifier is used, the anesthesiologist should be familiar with the potential hazards: overheating, leaks, “rain out,” changes in compression volume, and obstruction if connected in reverse. Humidifiers with servomechanism-controlled temperature regulation may prevent overheating and overhydration of infants. Anesthesia breathing circuits with internal heating elements may further reduce the rainout problem and minimize the risk of overheating; these advantages must be weighed against the additional expense. However, with the advent of warm air mattresses and more common use of circle systems in children of all sizes, there appears to be less of a need for in-line humidification as a means of warming children. The main indication for in-line humidification is to prevent secretions from drying when nonrebreathing circuits are used.

Wrapping

Any form of wrapping can reduce radiant and convective heat losses; plastic wrap used for food is particularly effective and inexpensive. Covering an infant’s head is of greatest value because the head represents such a large proportion of the body surface area. As with any warming device, there are hazards; the plastic drapes reduce conductive losses but also eliminate sweating as a mechanism of thermal regulation, so that hyperthermia may result.³⁴ “Space” blankets with reflective aluminized Mylar layers have been very effective in preventing heat loss in the past.³⁵ With effective room temperature control and forced air heating mattresses, the importance of wrapping the child has diminished.

Masks

The importance of elimination of mechanical dead space increases with decreasing size of the child. Rendell-Baker/Soucek masks, developed from molds of the facial contours of Caucasian children, were designed to minimize mechanical dead space without the inflatable cuff or high dome of adult masks. Transparent disposable plastic models are preferable to the classic black conductive rubber because they allow observation of a child’s color and condensate from exhaled humidity with respirations. Plastic disposable masks with soft inflatable cuffs around the periphery of the mask are particularly well suited for children with anatomic or mechanical problems that interfere with normal mask application. An especially useful mask has a built-in port, which allows passage of an endotracheal tube over a fiberoptic scope, thus providing excellent fiberoptic intubation conditions while maintaining spontaneous respiration, oxygenation, and depth of anesthesia (see E-Fig. 12-13).⁵⁴

Oral Airways

A complete selection of oral airways must be readily available. Infants have a relatively large tongue, which easily obstructs the airway once they lose consciousness. If too large an airway is inserted, damage to laryngeal structures (traumatic epiglottitis, uvular swelling) may result in postoperative airway obstruction (see Fig. 12-13).⁵⁵ Improperly inserted airways, by obstructing venous and lymphatic drainage, also may result in airway obstruction secondary to swelling of the tongue.^56,57 A tongue blade is useful to help insert the airway without kinking the tongue or catching the tongue or lips between the airway and the teeth.

Nasopharyngeal Airways

Nasopharyngeal airways are not as frequently used in children because the internal diameter is often small, resulting in increased work of breathing and increased chance of blockage by sections or blood. In addition, adenoid hypertrophy makes a child susceptible to bleeding after nasopharyngeal airway insertion. Despite its limitations, it can help provide a patent airway in spontaneously breathing infants and children (particularly after pharyngoplasty or cleft palate repair) and is better tolerated in awake children. Nasopharyngeal airways range in sizes to accommodate neonates through to adolescents and may have a movable ring to set the length or a flange on the end to prevent loss in the nasal cavity. If a stiffer airway is necessary, a well-lubricated tracheal tube, cut to the appropriate length (to reach the nasopharynx only), may be used; however, it must be safely secured because it may accidentally pass into the nasopharynx.

Laryngeal Mask Airway

The laryngeal mask airway (LMA) has proved to be a great addition to the airway management of children. Minimal practice is required to develop the skills for insertion. In some circumstances the LMA can be lifesaving (see Chapter 12).^58–64 This device generally provides a clear airway for maintaining spontaneous respirations; however, it cannot be relied on if controlled ventilation is required. Controlled ventilation has been described but holds the risk of gastric dilation.⁶⁵ We think that this device should be used primarily for spontaneous ventilation and with great caution with controlled ventilation in infants and children. Modern anesthesia ventilators provide the alternative and likely safer pressure-assisted mode of ventilation.⁶⁶ The LMA is particularly useful in children with anatomic airway abnormalities, because it provides a clear airway while other measures are planned for successful tracheal intubation or a tracheotomy.^64,67 Fiberoptic intubation through the LMA is possible in children with midfacial hypoplasia syndromes and in those with redundant airway tissue such as mucopolysaccharidoses (see also E-Fig. 12-22, A through C). Fiberoptic intubation is easier through the LMA because the child can be well oxygenated while using the fiberoptic scope to find the laryngeal inlet. Emergency placement of this device can also be lifesaving in children who are difficult to ventilate with bag and mask with or without a nasal or oral airway.⁶⁸

Variations to the classic LMA include disposable LMAs of all sizes, intubating LMAs in larger sizes, LMAs with reinforced tubing, and the ProSeal LMA (LMA, La Jolla, Calif.) with a distal opening that provides the ability to pass an orogastric tube (see Fig. 12-18, A and B). This variation to the classic LMA produces a better laryngeal seal and provides the ability to ventilate the lungs without a significant leak at greater peak inflation pressures.⁶⁹ This variation would seem to offer the greatest advantage in the child with a known difficult airway or in the emergency management of the child with a difficult airway who also has a full stomach. Many other LMA variants are available, each with claimed advantages. Each practitioner should assess which device is most effective in the practice environment⁷⁰ (see also Chapter 12 for a more in-depth discussion).

Tracheal Tubes

A variety of tracheal tubes must be available with appropriate sizes of stylets. There may be considerable variations from one manufacturer to another in wall thickness, external diameter, kink resistance, and direction and angle of bevel, as well as differences in tracheal tube cuff length and thickness of cuffed tracheal tubes.⁷¹ All tracheal tubes should be implantation tested and manufactured in accordance with the Z79 international agreement for safe products. Tracheal tubes are sized by the internal diameter (ID) in millimeters. Despite differences in endotracheal tube wall thickness among manufacturers, a first approximation for the correct tracheal tube size for an average child older than 2 years of age is⁷²:

Preterm infants weighing less than 1500 g generally require a 2.5-mm ID tracheal tube, whereas preterm infants of more than 1500 g require a 3.0-mm ID tube, full-term neonates require a 3.0- to 3.5-mm ID tube, a 1-year-old requires a 4.0-mm ID, and a 2-year-old requires a 4.5- to 5.0-mm ID. Although it has been suggested that the correct diameter of the tracheal tube may be determined by examining the diameter of the distal phalanx of the child’s first or fifth digit, evidence suggests that this is unreliable.⁶⁹ Tracheal tubes at least one half-size larger and smaller than estimated should be immediately available to accommodate airway size variability. The only true test for appropriate size selection is that the tube passes through the subglottis easily and has an appropriate leak (e.g., between 10 and 40 cm H₂O peak inflation pressure). The leak may be easily assessed by closing the circuit pop-off valve and slowly increasing the pressure by gently squeezing the anesthesia bag while listening over the larynx with a stethoscope. This technique has been demonstrated to be a sensitive and accurate measure of fit between the tracheal lumen and the endotracheal tube.⁷³ It should be borne in mind that a leak is dependent in part, upon the child’s head position and degree of paralysis.⁷⁴ For example, if the neck is flexed or the head is turned from side to side, a leak may become worse or decrease. If the trachea was intubated without the use of muscle relaxants, the child must be well anesthetized to assess the leak properly. If a child is coughing, it is difficult to make this assessment. Conversely, if the child has laryngospasm around the tube, then it will appear to be too large because a leak may not be present. On the other hand, too large a leak may prevent adequate ventilation should positive-pressure ventilation be required, pollute the OR excessively, and dilute the inspired gases during spontaneous respirations. In this circumstance, the tracheal tube should be replaced with the next larger size and the leak reassessed. The relationship of age, tracheal tube sizes and inhalaton pressure leak is commonly used to estimate laryngeal size (see Fig. 31-13).

The past decade has seen a transition from the routine use of uncuffed tubes in children during general anesthesia to the routine use of cuffed tubes. Uncuffed tubes are available for use from 2.0-mm ID to 6-mm ID and have been used as a standard in children up to 8 years of age. Indeed, traditional teaching suggests that the use of uncuffed tubes is the most appropriate practice⁷⁵; however, this is not evidence-based. The use of cuffed tracheal tubes has been demonstrated to reduce costs of inhalation agents use because of the ability to use lower fresh gas flows.⁷⁶ With the introduction of the MICROCUFF (Kimberly-Clark Health Care, Roswell, Ga.) tube and publication of several clinical studies, practice has shifted from the routine use of uncuffed to cuffed tracheal tubes. The external diameter of cuffed tracheal tubes is approximately 0.5 mm larger than uncuffed because of the cuff, so one must use a smaller-ID tube for a given tracheal size. Cuffed tracheal tubes as small as 3.0 mm ID are available for use in elective and emergency surgery in infants and children. Some have proposed that the advantages of a cuffed tracheal tube for children include reduced number of reintubations, decreased leak around the tracheal tube leading to less contamination of the operating room, greater ease and consistency of ventilation, reduced cost if low-flow anesthesia is used, and a theoretical attenuated incidence of aspiration.^76–82 If a cuffed endotracheal tube is used, the cuff is generally inflated just enough to provide a leak between 20 and 30 cm H₂O peak inflation pressure or cuff pressure can be monitored using a device that continually monitors cuff pressure.⁸³ It must recognized that the endotracheal tube cuff pressure will increase during a nitrous oxide (N₂O)-supplemented anesthetic because of diffusion of N₂O into the cuff,⁸⁴ unless saline is used to fill the cuff.⁸⁵ However, because the MICROCUFF tube seals the airway at a reduced pressure compared with polyurethane cuffed tubes, the time until the pressure within the cuff becomes excessive with the MICROCUFF tube is greater than with traditional tracheal tubes.⁸⁶ Excessive cuff pressures may present a special additional hazard to children, in whom swelling of the pseudostratified columnar epithelium that lines the inside of the cricoid ring may critically reduce the diameter of the upper airway. To date, studies have not investigated long-term complications such as subglottic stenosis that might occur in the intensive care unit. Tracheal tubes used in the OR remain in the trachea for only a brief time; thus it is unlikely that such use could result in long-term complications. An important advantage of the MICROCUFF tube is that the cuff is located closer to the tip of the tube (thereby reducing the potential for cuff pressure on the vocal cords or the cricoid cartilage, but eliminating the Murphy eye), lower cuff sealing pressure (again potentially reducing injury to the mucosa of the trachea), and less leakage around the cuff (less microaspiration) (see Fig. 12-15).^87–92 However, the marked nearly fivefold difference in cost will likely prohibit the routine use of this tracheal tube for short term intubation. Additionally, reports of easy kinking in the smaller size tubes resulted in a product recall.⁹³ The authors’ practice has increased the use of the Mallinckrodt Lo-Pro Endotracheal Tubes (Nellcor, Pleasanton, Calif.), which is not a new product but anecdotally provides a low-pressure cuff that has a lower contour, aiding the view during tracheal intubation, and has a smaller effect on the size of the outer diameter of the tracheal tube. No specific studies have demonstrated any differences in complications when comparing different low-pressure endotracheal tubes. The original Khine formula (4 + age [years]/4)⁷⁹ has been critically evaluated, and a new formula with greater tracheal tube size accuracy has been proposed (3.5 + age [years]/4).^79,94 However, these studies did not adequately examine the infants in the lowest age range, nor did they examine these specific brands of tracheal tubes. The argument could be made that long-term follow-up data for complications has not been adequately investigated.⁹⁵ Currently, we do not recommend the routine use of 3.0-mm ID cuffed tracheal tubes in neonates because the potential for laryngeal/tracheal damage far outweighs the advantage of avoiding an extra laryngoscopy and intubation.

Molded preformed tracheal tubes are especially useful for head and neck surgery because they remove the anesthesia circuit connections from the surgical field.⁹⁶ When excessive external pressure may be applied to the tracheal tube or when extreme flexion of the neck is likely to kink a standard endotracheal tube, a tracheal tube with Tovell spiral wire reinforcement may be used. Its kink resistance is well-known, but the additional wall thickness limits its applicability in younger children. The spiral springlike construction has caused the tube to pop out of the airway; in addition, repeated improper sterilization of nondisposable models may lead to bubble formation within the rubber, leading to airway obstruction when used with N₂O.⁹⁷

The carbon dioxide (CO₂) laser for treatment of laryngeal polyposis and other airway lesions has introduced the problem of ignition of the tracheal tube or other material in the airway. Avoiding high levels of O₂ and N₂O, which both support combustion, diminishes the risk of fire. Special stainless steel and metal implanted tracheal tubes are available but are quite expensive. Protecting the tube surface with aluminum or copper foil or wet sponges and red rubber tracheal tubes, which have a much reduced ignition potential, may offer the greatest protection (see Chapter 31).^98–102 Electrocautery can ignite standard tracheal tubes, esophageal stethoscopes, oral airways, feeding tubes, tissue, and packs if the correct circumstances combine.^103–105 Rubber tracheal tubes may soon no longer be manufactured because they are made of latex rubber but are still being advocated by some practitioners for laser surgery.

The tracheal tube itself creates significant resistance to airflow, and this phenomenon is described in the physics of resistance to flow for all gases and fluids. Flow of gases and fluids through airways and blood vessels and at branching points may be described as either laminar or turbulent. Reynolds number, a dimensionless value, is calculated as the ratio of inertial to viscous forces in the following equation:

where NRe is the Reynolds number, D is the diameter of the airway or blood vessel, and V is the velocity, ρ is the density, and µ is the viscosity of the gas or fluid. An NRe less than 2100 denotes flow that is laminar, whereas an NRE of 3000 or greater denotes turbulent flow. Flows that have NRe between 2100 and 3000 may be either laminar or turbulent. Conceptually, laminar flow is organized flow in which each plate of fluid, air, or gas moves in parallel at the same speed, with zero velocity at the interface between the wall of the airway or blood vessel and the gas or fluid. Laminar flow occurs in blood beyond the aortic arch and in airways beyond the fifth bronchial division. In contrast, turbulent flow is chaotic, unpredictable flow in which the fluid, air, or gas does not flow in parallel plates but rather develop eddies and lateral as well as forward flow, thus dissipating large amounts of energy. Turbulent flow occurs in the aortic arch and in the large airways of the tracheobronchial tree down to the fifth bronchial division.

Laminar flow is characterized by the Hagen-Poiseuille equation:

Where ΔP is the pressure drop, Q is the flow rate, η is the viscosity of the fluid (air or gas), l is the length of the airway or blood vessel, and r is the radius of the airway or blood vessel. Thus pressure drop is proportional to the flow rate and viscosity of the fluid and inversely proportional to the radius to the fourth power.

Turbulent flow is not easily characterized by an equation. Unpredictable losses such as friction are difficult to quantify. Hence, no single equation exists or turbulent flow as in the case of laminar flow. However, for large NRe, we know that:

Where ΔP is the pressure drop, Q is the flow rate, ρ is the density and r is the radius of the airway or blood vessel. Note that in contrast to laminar flow, pressure drop is proportional to the square of the flow rate and inversely related to the fifth power of the radius.

Therefore the resistance to gas or fluid flow is greater in relation to the longer and narrower tracheal tube (or intravenous catheter) used; anything that increases turbulence may proportionally increase the resistance to gas or fluid flow. In practical terms, this means that with a tracheal tube, when the respiratory rate is increased, turbulent flow results (meaning that resistance changes with radius to the fifth power, not the fourth power), decreasing the size of the delivered tidal volume.¹⁰⁶ In addition, flow may vary from laminar to turbulent at different portions of the respiratory cycle. Studies with in vivo models have shown that this effect is less pronounced with progressively larger tracheal tubes; however, larger tubes may result in reductions in delivered tidal volume when the respiratory rate is increased in children who require a smaller than normal endotracheal tube; that is, an increase in respiratory rate from 10 to 20/min does not double the delivered minute ventilation.^106–108 Data suggest that in children with poorly compliant lungs, lung compliance seems to be the most important factor, even with very small tracheal tubes.¹⁰⁶ The implications of these observations are that accurate measurement of end-expired CO₂ or arterial blood gases are the only way to be certain of the effectiveness of ventilation. In patients with severe pulmonary disease with ventilation-perfusion mismatch, only arterial blood gases or transcutaneous CO₂ monitoring will provide the needed information.

Routine Equipment

Intubation equipment must suit pediatric children of all sizes. Laryngoscopes are available with lightweight small handles and a full range of blades (see Chapter 12). An Oxyscope (Foregger, Inc. Langhorne, Pa.), which incorporates a small-bore O₂ delivery tube, reduces the incidence of cyanosis and bradycardia in spontaneously breathing neonates should laryngoscopy be prolonged (see E-Fig. 12-12, A and B).¹⁰⁹ Magill forceps are available in several sizes for children of different ages.

Special Airway Equipment

Difficult airways require special intubation equipment; flexible and rigid fiberoptic laryngoscopes and bronchoscopes are available for infants and children. Flexible fiberoptic laryngoscopes are available as small as 1.8 mm external diameter; these are suitable for passage of 2.5-mm ID endotracheal tube.¹¹⁰ The small scope is limited because it does not provide suction. Fiberoptic endoscopic skills should be developed during routine elective cases and in training exercises using airway or full manikin simulators to acquire these skills before encountering emergency cases or children with abnormal airway anatomy (see Chapter 12).¹¹⁰ Fiberoptic intubation through the LMA in infants and neonates with midfacial hypoplasia allows the maintenance of a clear airway, delivery of O₂ with or without inhalation anesthetic and provides optimal conditions for successful placement of the tracheal tube.^{61,62,111–114}

Lighted stylets provide an alternative technique for management of a difficult airway.^115–118 Other airway devices such as the Bonfils fiberscope (Karl Storz, Tuttlingen, Germany), the Airtraq disposable optical laryngoscope (Prodol Meditec, Vizcaya, Spain), the GlideScope video laryngoscope (Verathon, Bothell Wash.), the Storz DCI video laryngoscope (Karl Storz, Tuttlingen, Germany), and the Trueview PCD Infant (Truphatek, Netanya, Israel) can optimize the management of difficult pediatric airways^119–125 (see Chapter 12 for a more in-depth discussion).

Anesthesia Machine

All children can be anesthetized with standard adult machinery as long as the anesthesiologist is aware of the pitfalls and limitations. No anesthesia machines are currently marketed solely for pediatric use; however, anesthesia machines that have the ability to be customized for pediatric use may be of value, but not necessary. If one wishes to use special circuits, a Mapleson D system may be permanently mounted, allowing the choice of either a circle or open system. Modifying the anesthesia machine has inherent danger, especially for inexperienced users; the circuit or other components could be assembled incorrectly and potentially be unsafe. In addition, air should be available from a central pipeline or a cylinder with an appropriate yoke and flow meter. This is particularly advantageous in circumstances in which N₂O or a high inspired O₂ concentration is contraindicated (avoiding hyperoxia in a premature infant, N₂O in a patient with bowel distention, or laser airway surgery).^127–129 Even though machines are designed not to allow the delivery of a hypoxic gas mixture, an in-line O₂ analyzer and pulse oximetry are indicated to use an air–oxygen blend safely. The anesthesia machine should also be equipped with an alternative means of providing O₂ and positive-pressure ventilation (self-inflating bag) in case of a machine failure or malfunction.

Scavenger

The scavenging of anesthetic waste gases to limit exposure of anesthetic gases to health care providers is important. However, during pediatric anesthesia, exposure to anesthetic gases occurs during inhalational inductions by mask where high fresh gas flows are used and a poor mask seal exists because of patient anxiety or movement. The use of uncuffed tracheal tubes with large leaks also increases exposure. Modern scavenging systems are designed to avoid applying either negative or positive pressure to a patient circuit by using an open system, in which a reservoir is used to collect the waste gases and active (suction) or passive pressures are used to remove waste gases from the OR.¹³⁰ In a closed scavenging system, positive or negative relief values are used in the reservoir or waste gas exhaust hose to ensure that positive pressure does not build up in the system or if the system is blocked so negative pressure is not applied to the patient circuit.^131,132 In addition to scavenging systems, exposure to anesthetic gases is also related to adequate ventilation of the OR.¹³³

Circuits

A full spectrum of anesthetic circuits have been used in infants and children.¹³⁴ Adolescents and older children may be anesthetized with a standard adult semiclosed circle absorber system, perhaps with the substitution of a smaller rebreathing bag. Younger children may be anesthetized with the circuit modified by replacing the hoses with small-diameter pediatric tubing and by substituting a smaller reservoir bag. In the past, most infants and neonates (10 kg or less) were anesthetized with nonrebreathing open systems, such as the Mapleson D or F variety. These circuits are used infrequently today and may soon be of historical interest only because the increasing pressure for cost containment and the expense of the newer anesthetic agents has forced most children’s hospitals to convert to circle systems. In our institutions, we use only circle systems; this has simplified teaching and markedly reduced anesthetic agent costs. However, because Mapleson D systems are still being used, the advantages and disadvantages of open and closed systems for infants and neonates are reviewed in the following discussion.

Mapleson D systems do not have directional valves or a CO₂ absorber, thus eliminating the resistance intrinsic to the opening pressure of circuit valves and to turbulent flow through soda lime.¹³⁵ This may be an advantage in spontaneously breathing neonates during induction before intubation. The main disadvantage of open systems is the need for relatively high fresh gas flows and waste of expensive anesthetic agents, as well as unnecessary pollution of the environment. The work of breathing may be important for spontaneously breathing infants, so most pediatric circuits and masks are designed to eliminate both dead space and resistance. The classic example of this type of system is the Ayre T-piece; this circuit includes no valves or reservoir bags (Fig. 51-5).¹³⁶ Later modifications included changes to the T-piece and expiratory limb.¹³⁷ The Jackson-Rees modification involves the addition of a reservoir bag to the expiratory limb (Mapleson F).¹³⁸

FIGURE 51-5 The T piece as described by Ayre.

(From Ayre P. The T-piece technique. Br J Anaesth 1956;28:520-3.)

The Magill circuit and its various modifications, Mapleson A to E (Fig. 51-6) remain popular.¹³⁹ These systems consist of a source of fresh gas flow into the circuit, a reservoir bag, a pressure-relief (pop-off ) valve, tubing of various lengths connecting all of these parts, and an adapter for the mask or tracheal tube. Each of these has advantages and disadvantages, which are well reviewed elsewhere.^{130,134,140,141} The Mapleson D variety is the most commonly used because of its safety and versatility with both controlled and spontaneous ventilation.¹⁴¹ The pop-off valve is at the end of the expiratory limb, just before the reservoir bag; thus, fresh gas washes alveolar gas out of the expiratory limb during expiration. The efficiency of this washout (to prevent rebreathing) depends on the volume of the expiratory limb, the fresh gas flow, and the size of the tidal volume.¹³⁰ Larger children (heavier than 15 kg) require greater fresh gas flow rates to prevent rebreathing during spontaneous ventilation.^141,142 Fresh gas flow rates as low as 1 times minute ventilation may be used with controlled ventilation without an increase in expired CO₂ values.¹⁴³ Rebreathing (an increase in inspired CO₂) can occur in children with controlled ventilation when the respiratory rate increases above 20 breaths per minute; this apparent rebreathing can be diminished by using greater fresh gas flows (E-Fig. 51-2). The reason for this phenomenon may relate to inadequate washout of expired alveolar gas before the onset of the next breath. The advantages of minimal dead space and low resistance to flow with nonrebreathing circuits are counterbalanced by the disadvantages of heat and humidity lost to the anesthetic gases and significant waste of anesthetic agents because of the high flows required.^134,141 A clinically important factor is that whenever a change in anesthetic gas concentration is made at the vaporizer, this change is immediately reflected at the airway. This allows a more rapid induction of anesthesia, which may be an advantage but also means a greater risk of producing an anesthetic overdose when compared with a circle system. The most common and efficacious use of the Mapleson D circuit is for the safe transport of a child to and from the OR and for intrahospital transport. In this situation the light weight of the system, the ability to easily adjust peak inflation pressure and provide positive end expiratory pressure, and the ability to use relatively low O₂ flows make this system far superior to self-inflating (Ambu type) devices.

FIGURE 51-6 The Mapleson categories of breathing circuits. The most commonly used circuit for infants is the Mapleson D configuration.

(From Mapleson WW. The elimination of rebreathing in various semi-closed anaesthetic systems. Br J Anaesth 1954;26:323-32.)

E-FIGURE 51-2 Note the marked increase in inspired carbon dioxide values during anesthesia in a 7-kg infant using a Mapleson D circuit with 3 L/min fresh gas flow when the respiratory rate is increased from 10 to 20 and 20 to 40 breaths per minute. This may result from inadequate washout of alveolar gases between breaths. Increasing fresh gas flow or decreasing respiratory rate improves washout of alveolar gases. FG, Fresh gas; PIP, peak inflation pressure.

(From Rolf N, Coté CJ. Unpublished data.)

The circle system offers the advantage of using reduced fresh gas flows that decreases cost and decreases pollution of the atmosphere with waste anesthetic gases.^144,145 The use of the circle system in neonates and small infants generally requires assisted or controlled ventilation. Newer anesthesia machines use valves that require reduced pressure to open and close and the CO₂ absorbers are now wider and shorter, thus reducing the resistance. The main resistance to spontaneous ventilation in infants and neonates is the inner diameter and length of the tracheal tube. It is particularly important for practitioners using older systems with ventilators that do not compensate for circuit compliance or fresh gas flow to be aware of the different pressure and volume characteristics of circle systems compared with nonrebreathing systems (Fig. 51-7). Because most pediatric anesthesiologists recommend controlled ventilation in this age group, there no longer seems to be a great need for special circuits. The most important pitfall is that a tidal volume is selected based solely on weight rather than examining chest wall excursion, listening to breath sounds, observing the peak inflation pressure, and measuring end-tidal carbon dioxide (ETco₂) tension. Our in vitro studies have found that the main determinants of delivered tidal volume are peak inflation pressure, inspired to expired ratio, and respiratory rate.^106–108 As long as careful attention is paid to these important variables, even tiny infants can be successfully ventilated with adult circuits and adult bellows with pressure-limited ventilation (see further). No difference exists between adult circle systems and a Bain (Mapleson D) type system if pressure-limited ventilation is used and the same peak inflation pressure is generated.¹⁰⁸ The most important factor is that tidal volume per kilogram will be very large with an adult circle system compared with nonrebreathing systems, especially in infants. Many of these issues of delivering accurate small tidal volumes relative to the compliance of the circuit are simplified with the newer generation ventilators that compensate for the compliance of the system and the fresh gas flow (see later). In the final analysis, however, the practitioner must be cognizant of the fact that the only physiologic measure of the adequacy of ventilation is the CO₂ tension—ETco₂ or arterial CO₂.

FIGURE 51-7 The differences in compression volumes among five anesthetic circuits. The Mapleson D systems are the most efficient with the lowest compression volume, whereas the adult rubber circle system has the highest compression volume and is the least efficient circuit.

(From Coté CJ, Petkau AJ, Ryan JF, Welch JP: Wasted ventilation measured in vitro with eight anesthetic circuits with and without in-line humidification. Anesthesiology 1983;59:442-446.)

Carbon Dioxide Absorption

When using a closed or semiclosed circle system, removal of CO₂ from the circuit is required. This is accomplished by inserting a CO₂ absorption canister into the anesthetic circuit. The canister is usually placed between the anesthesia bag or ventilator and the fresh gas inlet. In some anesthesia machines, the canister contains two levels of a CO₂ absorbent in series. The replacement of the absorbent can create leaks in the circuit if the seal is not reapplied tightly,¹⁴⁶ complete obstruction if the plastic packaging is not removed from the prepackaged absorbent before insertion, or intermittent obstruction if part of the packaging is left in place.¹⁴⁷ The size of the absorbent is designed to maximize CO₂ reabsorption while minimally increasing the resistance in the circuit. Many different commercially manufactured absorbents are available whose chemistry involves the use of materials such as calcium hydroxide, sodium hydroxide, potassium hydroxide, and barium hydroxide, so that CO₂ is converted into calcium carbonate, water and heat. CO₂ absorption, cost and chemistry are reviewed in detail elsewhere.¹⁴⁸ The most common absorbents are soda lime and Baralyme (Allied Healthcare Products, Inc., St. Louis). Soda lime contains 95% calcium hydroxide, either sodium or potassium hydroxide and the balance as water. Baralyme has been voluntarily withdrawn from the market (see later). It contains 80% calcium hydroxide, 20% barium hydroxide, and the balance as water. Newer agents exist, such as Amsorb (Armstrong Medical Limited, Coleraine, Northern Ireland), which contains 70% calcium hydroxide, 0.7% calcium chloride, 0.7% calcium sulfate, 0.7% polyvinyl pyrrolidine, and the balance as water. Potassium hydroxide and sodium hydroxide have been incriminated in the degradation of halogenated anesthetic agents in CO₂ absorbers; thus the absence of these compounds in Amsorb prevents the degradation of inhaled anesthetics to carbon monoxide (CO) and compound A (see later). Most absorbents contain a dye such as ethyl violet that turns color as the pH of the absorbent decreases because carbonic acid is formed as CO₂ reacts with water. The change in the color in the absorbent indicates it is time to replace the absorbent.

Significant concern exists regarding the interaction of anesthetic agents with CO₂ absorbents because this interaction is an exothermic reaction that produces potentially toxic degradation products and a large amount of heat. Sevoflurane is degraded in the presence of CO₂ absorbent to produce five compounds, the most common of which is an olefin compound known as compound A.¹⁴⁹ Although compound A has been shown to cause renal corticomedullary necrosis in rats, extensive studies in humans have not yielded any substantive consequences or similar renal damage.^150–155 Factors that increase production of compound A include use of low-flow anesthesia, use of large sevoflurane concentrations, increased minute ventilation, anesthetics of prolonged duration, increased absorbent temperature, and fresh absorbent (Baralyme > soda lime Amsorb).^156–159 The safety of low-flow (less than 2 L/min fresh gas flow) sevoflurane is still a question because of the limited number of studies in children, although no anecdotal or animal evidence exists to suggest that children may be at greater risk with compound A.¹⁶⁰ Other toxic degradation products, including CO, formaldehyde, and methanol are produced when inhaled anesthetic agents interact with desiccated carbon dioxide absorbents (see Chapter 6). Carbon monoxide poisoning has been reported in patients with carboxyhemoglobin levels reaching 30% or more.^161–163 These case reports were often after several days of anesthesia machine inactivity, in particular, on a Monday morning (i.e., after a weekend with a large fresh gas left flowing through the machine all weekend while it was not in clinical use and usually without a reservoir bag attached).^160,161,164 In some anesthetic machines in which the fresh gas enters at or near the CO₂ canister rather than downstream of the inspiratory valve, fresh gas can dry the CO₂ absorbent by flowing retrograde through the absorber and exiting where the reservoir bag would usually be connected, because of lower resistance to retrograde flow when a reservoir bag is not in the circle system. In vitro studies demonstrated that CO production from the currently used inhaled agents occurs only when the CO₂ absorbent has been desiccated. The order of CO production from greatest to least is desflurane > isoflurane halothane and sevoflurane.^165,166 Other factors that increase the production of CO include the degree of desiccation and type of CO₂ absorbent (Baralyme > soda lime), increased temperature, increased anesthetic concentration, and low fresh gas flow rates.^{160,167–170} Newer CO₂ absorbents with little or no potassium or sodium hydroxide produce significantly less or no CO and compound A when they become dehydrated.^158,171,172 The Anesthesia Patient Safety Foundation developed a consensus statement in 2005 to address the problem of desiccation of CO₂ absorbents: “The APSF recommends use of carbon dioxide absorbents whose composition is such that exposure to volatile anesthetics does not result in significant degradation of the volatile anesthetic.”¹⁷³ The report discusses the difficulty and impracticality of monitoring CO₂ absorbent for desiccation and makes specific recommendations (if absorbents are used that degrade anesthetic agents), including turning off all fresh gas flow when the machine is not in use, changing the CO₂ absorbent frequently (e.g., every Monday morning), and changing the absorbent when there is concern about the state of hydration of the absorbent, such as when the machine is found with high fresh gas flows for an uncertain period.¹⁷³ New multiwavelength pulse oximeters are now available (e.g., Masimo Rainbow SET Pulse CO-Oximetry; Masimo Corporation, Irvine, Calif.) that have the ability to measure carboxyhemoglobin and methemoglobin levels noninvasively (see later discussion of multiple wavelength pulse oximetry); however, the costs of this technology make its routine use prohibitive.

A very important interaction of the modern halogenated anesthetics and desiccated CO₂ absorbent is a poorly defined exothermic reaction that has resulted in fires and explosions, specifically with sevoflurane and Baralyme. There have been several reports of extremely high heat in the CO₂ canister and inspiratory circuit in conjunction with fires, explosions, and various detrimental effects to patients, including burns, acute respiratory distress syndrome, and carbon monoxide poisoning (Fig. 51-8).^174–176 These cases often reported a very slow increase in sevoflurane concentration in the inspiratory circuit with a large gradient between the vaporizer setting and the measured inspiratory sevoflurane concentration along with a rapid color change of the CO₂ absorbent presumably from rapid degradation of sevoflurane and possibly a rapid change in pH or other substances causing the indicator color change. In 2003, Abbott Laboratories (North Chicago, Ill.), in conjunction with the U.S. Food and Drug Administration (FDA), released a practice alert and revised the product information about this phenomenon. This highly exothermic reaction occurs with other inhalation agents and other absorbents such as soda lime, but the highest temperatures of 392° F (200° C) and fires were associated with sevoflurane and desiccated Baralyme; Baralyme was voluntarily withdrawn from the market in 2004.^177–179

FIGURE 51-8 Sevoflurane/Baralyme interaction produced this fire in an anesthesia machine.

(Courtesy Mr. A. Rich.)

Humidifiers

The benefits of heating and humidifying inspired gases include maintenance of body temperature, improved mucociliary function, and prevention of drying secretions, particularly with nonrebreathing circuits.^180,181 A heated humidifier also helps maintain body temperature.^31–33 When using a circle system partial heat and humidification occur as part of the exothermic reaction used in the CO₂ absorber that adds heat and humidity to the circuit. Low fresh gas flows do increase the rebreathing of humidified and warmed gases, and when combined with passive humidification “artificial nose,” provides warm, humidified gases.¹⁵ However, small fresh gas flows add little heat and humidity in infants ventilated through a circle circuit.¹⁸² Humidification can be further increased by active humidification systems.¹⁵ Active humidification systems may create problems such as additional connections that may leak or become disconnected, “rain out” from humidified gas (resulting in water collection within the circuit or the tracheal tube and blockage of the capnograph), added resistance to gas flow, and bubbling noises that distract from the ability to auscultate heart tones and breath sounds. Inhalation of excessively heated inspiratory gases may result in “hot pot tracheitis.”¹⁸³ Whenever a heated humidifier is added, the extra gas-containing volume of the tubing and humidifier increases the total volume of the anesthetic circuit,¹⁸⁴ in turn increasing compression volume. The degree to which anesthetic circuit efficiency is affected depends on the gas-containing volume of the humidifier and the distensibility of the tubing used.¹⁸⁴ Because of significant complexity and potential for complications, these devices are no longer commonly used and likely soon to be of historic interest only.

Ventilators

First-Generation Anesthesia Ventilators

Nearly any adult-volume ventilator may be used with children by making appropriate adjustments to the respiratory rate, fresh gas flow, tidal volume, and inspiratory-to-expiratory time ratio. The use of first-generation adult ventilators in children, particularly neonates, requires meticulous attention to the settings before initiation of controlled respiration, to avoid barotrauma. In old anesthesia machines, most volume ventilators could be converted to pressure ventilators by adjusting the pop-off valve to the desired peak inflation pressure. When this modification is made, special attention must be paid to the peak inflation pressure because a fractional turn of the pop-off valve in either direction could result in either excessively high or inadequate peak inflation pressures.

The next generation of anesthesia machines, which have a switch at the CO₂ canister to change from manual to mechanical ventilation, do not allow adjustment of peak inflation pressure at the pop-off valve because in the mechanical ventilation mode the pop-off valve is out of the circuit. When such a configuration is used, the pop-off valve of the ventilator must be used to limit the peak inflation pressure. The excursion of the ventilator bellows may also be adjusted to limit the size of each tidal volume. Additionally, limiting the inflow of gas to the bellows will limit the bellows inflation. Any of three techniques for adjusting the excursion of the adult bellows provides equal tidal volume as long as the anesthesiologist pays careful attention to peak inflation pressure.^106–108 Using the adult ventilator in pressure pop-off mode allows safe use even in neonates with a very small tidal volume. The following is a suggested means for setting up an older adult ventilator for use with children in the pressure-controlled mode:

1. Set pop-off limit on ventilator to 20 cm H₂O.

2. Set rate to appropriate age.

3. Set inspiratory to expiratory ratio to 1 : 2.

4. Set gas flow to the middle range.

5. Set tidal volume to a minimum of 10 mL/kg or minute ventilation (VE) to a minimum of 200 mL/kg/min.

6. Turn lever to ventilator mode (new systems) or attach ventilator hose to replace bag (old system).

7. Observe peak inflation pressure to be certain that it is 20 cm H₂O or less. If the peak inflation pressure is less than 20 cm H₂O, recheck pressure pop-off limit to make sure it is set at 20 cm H₂O, and, if correctly set, slowly increase the size of the tidal volume until 20 cm H₂O is achieved.

8. Auscultate both lungs and observe chest excursion and ETco_2.

9. Make certain fresh gas flow is adequate to maintain bellows at full capacity.

10. Reassess the entire process based on Step 8.

11. For noncompliant lungs or with a small-diameter tracheal tube (e.g., 2.5 mm ID), greater peak inflation pressures may be required.^106–108

It is our bias that manual bag ventilation provides immediate feedback about a child’s lung compliance or problems with the circuit (e.g., kinked tube, disconnections, or bronchospasm). However, the literature is unclear regarding this issue. Some experienced anesthesiologists have been able to detect an occluded endotracheal tube, whereas others have not.^185,186 This may in part depend on fresh gas flow, the size of the child, and the configuration of the circuit. The use of a ventilator provides valuable assistance by freeing the anesthesiologist’s hands for other use when the immediate hand-bag feedback is given up for the convenience of a mechanical ventilator. A disconnect, apnea, or peak inflation pressure alarm system is helpful to aid vigilance for these types of adverse events, although these alarms are not perfect.¹⁸⁷ A CO₂ detector with the alarm enabled is optimal; the American Society of Anesthesiologists’ (ASA) Standards for Basic Anesthetic Monitoring require “continual monitoring for the presence of expired CO₂” and the “alarm shall be audible to the anesthesiologist or the anesthesia care team personnel.”¹⁸⁸

With many first-generation ventilators, fresh gas flow may have considerable impact on tidal volume during volume-controlled ventilation¹⁸⁹; fresh gas flow into the circuit is added to the ventilator output during the inspiratory time. This augmentation effect may result in serious errors in calculating delivered VE during volume-controlled ventilation. The following examples illustrate this situation in volume-controlled ventilation:

1. An adult has a VE of 7 L/min, I : E ratio of 1 : 2, and fresh gas flow of 6 L/min. If there were no compliance or compression volume losses, the patient would receive 7 L/min + ( × 6 L/min) = 9 L/min. If the fresh gas flow were reduced to 3 L/min, the patient would receive 7 L/min + ( × 3 L/min) = 8 L/min. This is a change of about 11%.

2. A child has a VE of 2 L/min, I : E ratio of 1 : 2, and fresh gas flow of 6 L/min. Again, assuming no compression or compliance volume losses, the child would receive 2 L/min + ( × 6 L/min) = 4 L/min. If the fresh gas flow were now reduced to 3 L/min, the child would receive 2 L/min + ( × 3 L/min) = 3 L/min. This is a 25% decrease in delivered minute ventilation (E-Fig. 51-3). Compared with the earlier example in adults, the impact on the minute ventilation in a child is magnified 2.5-fold.

Thus when using first-generation anesthesia ventilators, increases and decreases in fresh gas flow during volume-controlled ventilation result in changes in delivered VE. The potential for clinically important effects on ventilation is greatest in infants (e.g., several study patients were found to have a 40% difference in VE when fresh gas flow was changed from 1.5 L/min to 6.0 L/min without any change in the ventilator settings.¹⁹⁰ When using volume-controlled ventilation, if a change is made in fresh gas flow for any circuit used in children, the adequacy of chest expansion, breath sounds, peak inspiratory pressure, and ETco₂ must be reevaluated, particularly when using older ventilators that do not automatically compensate for such changes (see later discussion of new ventilators).¹⁹¹ This effect does not occur during pressure-controlled ventilation because excess flow is simply diverted out of the pop-off valve. However, if fresh gas flows were reduced such that the tidal volume was less than the peak pressure setting, the tidal volume delivered to the child could be reduced.

E-FIGURE 51-3 Changes in delivered minute ventilation using a pediatric circle system, as reflected by end-tidal carbon dioxide (CO₂) measurement, in a 6-kg child during volume controlled (not pressure controlled) ventilation. The only change made was in fresh gas flow. Note the near halving of end-expired CO₂ when fresh gas flow was increased from 1.5 L/min to 6 L/min. Whenever a change is made in fresh gas flow during controlled ventilation, new ventilatory settings must be determined. PIP, Peak inflation pressure; RR, respiratory rate; V•E, minute ventilation.

The Anesthesia Record and Bar Code Drug Administration

In addition to its obvious role as a medicolegal document, the anesthetic record can be a very important monitor. Proper recording of a child’s status on arrival in the OR encourages evaluation of the effects of any premedication, confirmation of nil per os (NPO) status, allergies, current medications, and assessment of weight and fluid balance. Careful recording of intraoperative fluid administration and losses allows assessment of a child’s fluid replacement. Concurrent charting of vital signs and anesthetic drugs administered allow correlations to be made and encourages trend analysis. Many changes that are too subtle to interpret on a moment-to-moment basis become obvious when graphed out over time. In addition, a numbering system correlating events with time on the anesthesia record documents the sequence of anesthetic management and may prove very useful should a medicolegal issue arise.

Automated anesthesia records or automated anesthesia information management systems (AIMS) are increasingly used and will eventually replace hand-recorded records and likely improve the accuracy of data collection.^223–226 These systems can provide timely reminders for administration of antibiotics, documentation of start and stop times, compliance with regulatory requirements, reminders of blood pressure measurement gaps, and automatically provide cumulative totals in infusions of drugs, such as propofol or remifentanil.²²⁷–²³⁰ Most importantly, the automated recording of physiologic data in real time provides improved accuracy, particularly at times when the anesthesiologist is multitasking or when an emergency occurs. In fact, such data may actually be protective from medical malpractice because the veracity of the data is far improved over retrospective data recording.^231,232 Conversely some data recorded are spurious and may in fact be nonreflective of an actual event; neither system is perfect.^233,234 Currently, the widespread introduction of automated record-keeping has been very slow for several reasons, including the large capital and contract expenses for implementation, difficulty integrating all hospital-wide digitized systems in real-time terms, and costs for 24-hour support staff.²²⁵

Medication errors are quite common, particularly in high acuity areas such as the OR and the ICU.²³⁵ The next area of technology improvement may be the adoption of bar code confirmation of drug administration in the operating room.^236–242 Such systems are currently available to provide accurate labeling of syringes (date, time, concentration) and have had great success in reducing medication errors.^{235,243–251} Several studies have demonstrated reduced medication errors in the OR when bar coding is used.^236,242,251 These systems will be incorporated into the automated anesthesia record in the future as well.

Precordial and Esophageal Stethoscope

In infants, neonates, and small children, an experienced ear can easily detect arrhythmias but also may be able to assess changes in cardiac output and blood pressure (soft heart tones compared with baseline).²⁵² The continuous use of a stethoscope is of particular value in situations in which more advanced noninvasive blood pressure monitoring (NIBP) is unavailable. The precordial stethoscope may be stabilized with a double adhesive disk. For specialized needs, such as bronchography, angiography, or magnetic resonance imaging, a plastic stethoscope that is not radiopaque or ferromagnetic may be used. The optimal site where both heart tones and breath sounds can be heard is usually at the apex of the heart, but occasionally the suprasternal notch provides better listening conditions. The latter position may be more advantageous during induction and emergence, because information regarding airway patency is more readily obtained. The stethoscope can provide early indications of developing airway obstruction or laryngospasm, allowing the practitioner to take corrective action (PEEP) before a full-blown episode of laryngospasm develops. However, the perception is growing among clinicians that current monitors have superseded the need for the precordial stethoscope.²⁵³

Whenever the trachea has been intubated, the precordial stethoscope may be exchanged for an esophageal stethoscope. The pediatric size esophageal stethoscope may be introduced atraumatically, even in neonates. The optimal position for the stethoscope within the esophagus can be ascertained by inserting the scope while listening for the position that provides the loudest heart and breath sounds. The less expensive adult variety is often usable in children age 6 years or older. Some disposable esophageal stethoscopes incorporate a thermistor, allowing the introduction of two monitors at once. One possible relative contraindication to using or inserting an esophageal stethoscope is during a tracheostomy. Misidentification of an esophageal stethoscope as an endotracheal tube has resulted in the surgeon opening the esophagus.²⁵⁴ During a tracheostomy in a child, either the esophageal stethoscope should not be used or the surgeon should be informed that two stiff tubes pass through the neck structures, because this complication has occurred even in the hands of very experienced surgeons.²⁵⁵

The major drawback of a stethoscope as a monitor is that it provides information only when it is connected to the anesthesiologist. Custom-molded earpieces, which better exclude room noises, are much more effective and more comfortable to wear than the conventional binaural stethoscope. These earpieces have the added advantage of leaving one ear open for communication with the surgeon and ancillary personnel. Even with pulse oximetry and capnography the stethoscope can provide great reassurance that the patient has an adequate blood pressure and cardiac output. If both the NIBP and the oximeter fail, but the child has strong heart tones, a technical problem likely exists. However, if the other monitors fail and the heart tones are very weak, then cardiac output may be compromised and attention should be immediately focused on resolving that problem rather than wasting time trying to search for a problem with the monitor.

Blood Pressure Devices

Blood pressure must be monitored in every child who requires our care. An appropriate-size blood pressure cuff should cover approximately two thirds of the length of the upper arm or thigh.^256,257 The cuff bladder should rest over the artery; in older children a stethoscope may be secured over the artery to listen for Korotkoff sounds. In smaller children, the flicker of the needle in an aneroid sphygmomanometer dial may be used as an indicator of systolic pressure.²⁵⁸ If the flicker is not clearly visible, a distal pulse sensor may be used. Such a sensor can be either a photoelectric plethysmograph on a digit that detects the return of the signal as the cuff deflates or a Doppler flow detector that is positioned over a distal artery.²⁵⁹ Electronic oscillometer units (NIBP) are capable of repeated, frequent, and accurate measurements.^260–263

If an automatic NIBP device is used, it is important that proper application, function, and adequate deflation time be ensured; venous stasis, petechiae, and nerve compression damage are possible, especially with early models.²⁶⁴ Most automated NIBP machines have specific cuff sizes for infants, neonates, and preterm infants, as well as a requirement that the monitor settings be matched with the size of the patient (neonate, child, adult). For neonates and infants, special tubing is required to connect the cuff with the monitor. If cuffs are improperly matched with the tubing and monitor settings, the built-in algorithms will produce factitious data.²⁶⁵ It appears that these devices offer reasonable accuracy for systolic blood pressure but may be less accurate for diastolic blood pressure.^266,267 NIBP monitors provide very useful information during induction of anesthesia, particularly important during that period before establishing intravenous access. The results of one study showed greater consistency and accuracy with these devices than with standard auscultatory methods.²⁶³ One common practice is to place the cuff on the calf of infants as a substitute for upper extremity measurements. There is poor correlation between arm and calf blood pressures; because of this variability, one should not simply assume that one site is a substitute for the other.^268–271 Another study determined that the loss of the pulse oximeter waveform correlated with systolic arterial pressure in children weighing less than 15 kg.²⁷²

Electrocardiograph

Although many arrhythmias may be diagnosed by careful attention to the heart sounds conducted through the precordial or esophageal stethoscope, the electrocardiogram (ECG) remains a mandatory monitor for all instances of sedation, regional anesthesia and general anesthesia. In a healthy child, lead placement may differ from that selected for adults because the principal need is for diagnosis and resolution of arrhythmias rather than detection of ischemia. One report emphasized that children undergoing cardiac surgery or children with severe anemia may develop ST depression consistent with ischemia, although it occurs extremely rarely.²⁷³ With the greater right heart predominance in the younger ages, a cross-chest lead generally provides the optimal combination of atrial and ventricular voltage signals. A QRS detector with a beeper is a useful accessory, particularly when vagal stresses occur. The ECG heart rate is used for comparison with the pulse oximeter heart rate to confirm appropriate application and function of the pulse oximeter. A built-in lead fault detector is also useful to check the cables and a patient’s leads when a poor signal is obtained. Cleaning, gently abrading, and defatting the skin with alcohol can improve the contact. Tincture of benzoin may also help maintain secure adhesion in areas near the application of surgical preparation solutions. Care must be taken not to allow the leads to become wet with preparation solutions and to isolate the leads from the electrocautery dispersive electrode to avoid electrical burns.²⁷⁴ Special ECG leads and monitors are required for use in the MRI scanner (see Chapter 45).

Oxygen Monitor

Assessment of the inspired O₂ concentration is one aspect of the monitoring guidelines published by the ASA and Harvard Medical School.^193,275 It is necessary in preterm infants and neonates to be alert to the possible relationship between high arterial O₂ tension (Pao₂) resulting from high inspired O₂ concentrations and ocular toxicity, although recent evidence suggests that this relationship is not a quid pro quo.²⁷⁶ This monitor is also helpful in guiding inspired O₂ concentrations in certain types of congenital heart disease in which low inspired O₂ is key to optimizing pulmonary artery pressures and pulmonary blood flow. An O₂ monitor allows easy blending of air and O₂ to achieve the desired inspired O₂ concentration. In all children, O₂ monitoring can help prevent hypoxemia. The standard, so-called fail-safe device built into most anesthesia machines is generally keyed only to pressure in the O₂ line; should the O₂ flow be turned off, a child may receive a potentially hypoxic mixture (air plus a large concentration of inhalational anesthetic such as sevoflurane or desflurane) without warning from the fail-safe alarm. An O₂ monitor on the inspiratory limb would indicate a decrease in the inspired fraction of O₂ and provide an accurate alarm. With the more frequent use of closed-circuit anesthesia and air–O₂ blends, O₂ monitoring assumes even greater importance. Anesthesia machines generally provide a mechanical or electronic coupling system that delivers minimal O₂ flow and prevents the delivery of hypoxic mixtures of N₂O and O_2. However, some newer machines do not have a minimum O₂ flow and disable their electronic coupling system when air and O₂ are combined; thus it is possible to intentionally or unintentionally turn on the air without O₂ to deliver 21% O₂. The delivery of low inspired O₂ is sometimes desirable in certain conditions such as hypoplastic left heart syndrome or to decrease the risk of airway fire when laser or electrocautery are being used. When high concentrations of sevoflurane or desflurane are used with air it is possible to dilute the O₂ and deliver a potentially hypoxic mixture. Careful monitoring of inspired O₂ is vital to prevent this from occurring unintentionally. A high inspired O₂ alarm device is also desirable; this may detect when an air tank is empty. The O₂ analyzer should be calibrated and alarm limits set before anesthetic induction; a narrow band of alarm limits allows early detection of changes in gas flow ratio.

Temperature Monitors

Although there is medicolegal pressure to monitor temperature in every child because of the potential danger of malignant hyperthermia, in reality temperature monitoring is mandated for other reasons. Hypothermia is much more common in the OR than hyperthermia and may be associated with acidosis, myocardial irritability, coagulopathy, respiratory depression, prolonged neuromuscular blockade, greater absorption of inhalation agents, and delayed emergence from anesthesia.^277–280

NonInvasive Oxygen Saturation Monitors

Pulse Oximetry

Nothing has changed anesthetic practice more in the past few decades than the introduction of mandatory pulse oximetry for all cases requiring sedation and general anesthesia. This monitor determines the O₂ saturation of hemoglobin by spectrophotoelectric oximetric techniques. Measuring the amount of light transmitted through tissue between a two-wavelength light source (930 and 660 nm) and a detector allows continuous calculation of arterial O₂ saturation.²⁸¹ The use of two frequencies helps eliminate interfering absorption by other molecules, although intravenous dyes (indocyanine green, methylene blue) and colored nail polish may interfere with normal sensor function and cause the oximeter into reading a factitiously low O₂ saturation value.^275,282,283 Additionally, dyshemoglobinopathies such as carboxyhemoglobinemia and methemoglobinemia may result in significant artifact. Carboxyhemoglobinemia causes an overestimation of O₂ saturation because the photodetector incorrectly interprets carboxyhemoglobin as oxyhemoglobin; this artifactual change is roughly proportional to the concentration of carboxyhemoglobin.²⁸⁴ Methemoglobinemia results in desaturation, but the saturation recorded tends to read greater than the actual saturation; at high levels of methemoglobin during episodes of desaturation, this disparity becomes greater.²⁸⁵ Fetal hemoglobin, hyperbilirubinemia, and sickle cell disease have minimal effect on pulse oximetry.^286–290 This monitor is easy to use, noninvasive, and reasonably accurate, although not perfect.^291,292 Of greatest importance to the clinician is appreciating that the current oximeter technology underestimates the actual hemoglobin saturation as the saturation is decreasing; the more rapidly the saturation decreases, the more the saturation reading underestimates the true saturation.

The clinical efficacy of this monitor was demonstrated in a prospective single-blind study of pediatric patients; 50% of the cases were conducted with the saturation data and alarms made available to the anesthesia team, whereas in the remainder, the data were blinded and the alarms silenced. Several observations were reported^293,294:

1. Twice as many “major” desaturation events (defined as saturation 85% or less for 30 seconds or longer) occurred in the blinded group.

2. The oximeter detected a greater number of desaturation events before clinical recognition by the anesthesiologist.

3. Major desaturation events were not accompanied by changes in vital signs in most cases (i.e., in only 4 of 19 such events was any change in blood pressure, heart rate, or respiratory rate noted).

4. The incidence of these events in ASA physical status 3 and 4 patients was greater than in ASA 1 and 2 patients.

5. Desaturation events occurred in children who were managed by both experienced and inexperienced anesthesia personnel.

6. Correlation between the observation of cyanosis and true desaturation was poor.

7. Desaturation events were as likely in brief as in procedures of extended duration.²⁹³

In a subsequent study, the efficacy of combined pulse oximetry and capnography was examined to determine if patient safety was improved with the addition of capnography.²⁹⁴ In 400 children, 260 problems were observed (E-Fig. 51-4). The study confirmed that pulse oximetry was superior to the human eye in diagnosing hemoglobin desaturation. The incidence of major desaturation events when the oximeter data and alarms were blinded from the practitioners was threefold greater than when they were available. Capnography was not as helpful in diagnosing the majority of events leading to hemoglobin desaturation (E-Fig. 51-5). This is consistent with the Australia anesthesia incident study in which the authors reported that oximetry provides useful data in 85% of events.²⁹⁵ In the pediatric study, 15 problems fulfilled the criteria of a major capnograph event. These were problems that posed a threat to life (e.g., kinked tracheal tube, circuit disconnect, esophageal intubation, accidental extubation) and would have been detected immediately by a capnograph, but 8 were diagnosed initially by pulse oximetry (E-Fig. 51-5). Capnography may have helped initially to diagnose the seven remaining events; however, pulse oximetry would have diagnosed these events as desaturation developed, 30 to 60 seconds later. This observation is also consistent with the Australia incident study.²⁹⁶ This study also examined the incidence of “minor” capnograph events, defined as an abnormality of ventilation that was not a threat to a child’s life (e.g., hypercarbia or hypocarbia). A significantly greater incidence of both hypercarbia and hypocarbia occurred in children whose anesthesiologist was blinded from the capnograph data and alarms.

E-FIGURE 51-4 A total of 260 problems were observed in 153 of 402 patients studied. Major desaturation events (SpO₂ ≤85% for ≥30 seconds) and major capnograph events (esophageal intubation, kinked endotracheal tube, circuit disconnect, accidental extubation) are within the bold circles. Minor desaturation (SpO₂ ≤95% ≥0 second) and interesting events (SpO₂ ≤85 or ≥30 seconds) are also presented. Minor capnograph problems were simple hypercarbia (PECO₂ ≥60 mm Hg ≥60 second) or hypocarbia (PECO₂ ≤25 mm Hg ≥60 seconds). The total number of events that fulfilled the criteria for each category is in parentheses. Note that some events fulfilled criteria in two or more categories. PECO₂, Expired carbon dioxide; SpO₂, oxygen saturation.

(From Coté CJ, Rolf N, Liu LM, et al. A single-blind study of combined pulse oximetry and capnography in children. Anesthesiology 1991;74:980-7.)

E-FIGURE 51-5 Fifty-nine major desaturation events were observed. Nearly 70% were first diagnosed by oximeter, 22% by the anesthesiologist, and only 8% by the capnograph. Pulse oximetry is by far the more sensitive monitor in providing an early warning of developing desaturation.

(From Coté CJ, Rolf N, Liu LM, et al: A single-blind study of combined pulse oximetry and capnography in children. Anesthesiology 1991;74:980-7.)

Infants 6 months of age or younger are at greatest risk for experiencing at least one major desaturation event (E-Fig. 51-6) or a major capnograph event. The number of children with multiple problems in the group in which neither pulse oximetry nor capnography data were available to the anesthesia team was twice as great as in the group in which both were available. An interesting observation was the relatively high incidence of clinically unrecognized endobronchial intubation that manifested as persistent, though minor, hemoglobin desaturation (93% to 95%).²⁹⁷

E-FIGURE 51-6 The pie figure on the left presents the age distribution of the 402 patients studied, and the pie on the right presents the number of patients in each category with a major desaturation event. Note that infants had a significantly higher incidence than did toddlers or children (P = 0.001).

(Modified from Coté CJ, Rolf N, Liu LM, et al. A single-blind study of combined pulse oximetry and capnography in children. Anesthesiology 1994;74:980-7.)

These studies confirm that the pulse oximeter provides an early warning of developing desaturation well before a clinician is able to detect it clinically. This is easily explained by the fact that approximately 5 g of desaturated hemoglobin is required to detect cyanosis; if a child’s hemoglobin level is 15 g/dL, the O₂ saturation would have to decrease to 66% (i.e., less than the normal venous hemoglobin desaturation before cyanosis would be clearly evident). Pulse oximetry detects evolving hemoglobin desaturation in advance of this value being achieved. There is some variability among manufacturers, and accuracy declines as the saturation values decrease (70% or lower).^284,298 The degree of inaccuracy of pulse oximeters is a function of many variables that should be carefully reviewed.^286,299,300 Pulse oximetry is a reliable monitor for the majority of children who are free of cyanotic congenital heart disease.

Quite apart from the hemoglobin O₂ saturation data, pulse oximetry provides additional information including data pertinent to the systemic blood pressure.²⁷² Pulsus paradoxus may be observed in hypovolemic children. Loss of the peripheral pulse when using pulse oximetry may accompany hypovolemia, vasoconstriction resulting from hypothermia, or inadequate cardiac output as a result of hypovolemia, allergic reaction, or anesthetic overdose.^301,302 Pulse oximetry is useful as a noninvasive trend monitor in individuals with anatomic or physiologic shunts, because the effects of anesthetic agents and positive-pressure ventilation are difficult to predict.^303–305 Pulse oximetry also can be an effective monitor for preventing hyperoxia in preterm infants by guiding the inspired O₂ concentration to values that correspond to hemoglobin saturations between 93% and 95%.³⁰⁶ Preductal and postductal oximeters indicate opening and closing of a patent ductus arteriosus (PDA).

Various common OR events can affect the accuracy and reliability of pulse oximetry including electrocautery, flickering operating room lights, movement, blood pressure cuff inflation, bright light, vasoconstriction, stereotactic positioning systems,³⁰⁷ and injected dyes (cardio green and methylene blue). Falsely high O₂ saturation measurements may occur despite severe hemoglobin desaturation in the presence of interference with the oximeter sensor. Darkly pigmented skin color may also affect accuracy with up to a 10% error in low saturation ranges.³⁰⁸ Consequently, it is important to remember that this monitor is not foolproof.³⁰⁹ One very important artifact, the penumbra effect, is caused when an adult clip-on sensor is placed on an infant’s finger or toe.³¹⁰ In this situation the oximeter senses the motion of the pulse but light is transmitted around the finger or toe, thus providing a falsely high hemoglobin saturation. Only appropriate-size oximeter probes should be used in infants and children. In critically ill children and those with thermal injuries, problems with finding a suitable site to apply the probe can be overcome by using a modified oximeter probe applied to the tongue (see E-Fig. 34-5, A to D).^311,312 The tongue provides a rich blood supply unaffected by vasoconstriction, and electrocautery interference is less. If the tongue moves, the sensor may dislodge or record a false heart rate. Thermal injury secondary to the interface of incompatible sensor probes and cables (i.e., equipment from more than one manufacturer) has been reported, as has pressure necrosis resulting from too tight an application.^313,314

Recent advances in pulse oximetry technology have attempted to make several major improvements, including (1) software that provides motion artifact compensation; (2) software that allows uninterrupted readings even in low flow states; and (3) software that provides more accurate assessment of oxygenation in cyanotic patients.³¹⁵ Conventional pulse oximetry time-averages a signal over 5 to 10 seconds, depending on the default settings of the oximeter. This signal represents the ratio of transmitted red and infrared light, and the instrument uses this to determine the child’s hemoglobin saturation. This averaging smoothes motion-induced artifacts but may ultimately lessen the sensitivity and response time of a conventional oximeter in situations of hypoxemia or even delay the recognition of asystole or severe bradycardia by several seconds. More recent models are designed to recognize motion artifact and discard signal aberrations secondary to motion using proprietary signal-filtering techniques. Some devices combine two different approaches to signal processing to arbitrate the correct saturation based on dual signal filtration techniques using both pattern matching and adaptive comb filtering. This technique is designed to detect motion and reject analysis of data generated during motion. Other oximeter manufacturers take a different approach, using spectrophotometer principles to generate a saturation signal. That saturation signal may contain several component saturations corresponding to different absorption artifacts of blood and other tissues, including venous blood. The signal is then processed using a discrete saturation transform (DST) algorithm that then extracts the highest and presumably the arterial component saturation and reports this value as the hemoglobin O₂ saturation.^316–319 Conventional pulse oximeters can overestimate or underestimate the degree of desaturation and therefore may have less accuracy and greater signal dropout rate (i.e., no reading or poor quality signal associated with an alarm, especially in children with hypoxemia). These deviations may be improved with the newer generation pulse oximeters, but this has yet to be definitively demonstrated.³²⁰ The newer generation pulse oximeters perform better than conventional pulse oximetry in hypothermic low flow states during cardiopulmonary bypass.³²¹ Movement artifact is the main cause of false alarms in sedated or awake children who require oximetry monitoring. This artifact is significantly reduced with the current generation of pulse oximeters in both adult and pediatric patient populations.^322–326

Reflectance Oximetry

Reflectance pulse oximetry has been available for many years, but with recent technologic improvements, several devices have received FDA approval and are currently marketed. Reflectance oximetry technology was first created to improve O₂ saturation data in difficult clinical conditions, such as decreased perfusion, movement, and conditions in which optimal sites for standard transmission oximetry probes may not exist (e.g., burns). The technology behind reflectance oximetry involves the emission of multiple wavelengths of red and infrared light that are sensed in two or more photodetectors located a distance away from the diodes on the same probe sensor (Fig. 51-9). These devices measure reflected rather than transmitted light. The technology used to create these oximeter probes permit usage on the forehead, in the esophagus, and on the fetal scalp. The forehead devices have had the largest use among pediatric anesthesiologists. Studies of reflectance oximeters in general demonstrated that these devices are more susceptible to erratic measurements when the probe is placed directly over an artery³⁴⁴ and vein.³⁴⁵ Some potential advantages to using reflectance oximetry on the forehead include better signals during conditions of poor perfusion, lack of motion, and faster response time than conventional peripheral oximeter probes.^346,347 However, possible disadvantages include a potentially high signal dropout rate (no measurable signal) and poor accuracy, often measuring lower than finger transmission oximetry, particularly in cases of venous congestion (e.g., Trendelenburg position or situations that impede venous return).³⁴⁸ Several studies, including more recent studies, report good correlation between concurrent measurements of O₂ saturation comparing reflectance oximetry on the forehead,^349–351 the esophagus,³⁵² and around the chest of premature infants,³⁵³ compared with extremity pulse oximetry and arterial co-oximetry measurements; these studies also report a clear dropout rate.

FIGURE 51-9 A, Comparison of conventional transmission pulse oximetry with reflectance oximetry. Note that with conventional oximetry the light detector is located directly opposite on the other side of a digit and detects light transmitted through the digit; with reflectance oximetry the light detector is located next the light emitters and detects the light reflected back through and from the tissues. B, The typical configuration of a reflectance oximeter. C, Another configuration, in which several light detectors are located at different distances from the light emitters. In addition, reflectance oximeters may emit light in more than just 2 wavelengths, which is typical of conventional transmission pulse oximeters.

(Modified from Kugelman A, Wasserman Y, Mor F, et al. Reflectance pulse oximetry from core body in neonates and infants: comparison to arterial blood oxygen saturation and to transmission pulse oximetry. J Perinat 2004;24:366-71; and Keogh BF, Kopotic RJ. Recent findings in the use of reflectance oximetry: a critical review. Cur Opin Anesth 2005;18:649-54.)

Near-Infrared Spectroscopy

Near-infrared spectroscopy (NIRS) is a noninvasive, optical technology that bears similarities to pulse oximetry in that it uses the relative absorption of near-infrared light, 700 to 900 nm, through biologic tissues to determine tissue oxygenation (E-Fig. 51-7).³⁵⁴ Oxyhemoglobin and deoxyhemoglobin absorb light at different frequencies, and the use of a probe that emits and detects different frequencies of near-infrared light can be used to estimate tissue oxygenation. This device has been most widely used to measure “regional” cerebral O₂ saturation (rSO₂),^355,356 but measurement of oxygenation in other areas, such as kidney, bowel, spinal cord, free tissue flaps, and other areas, is possible.^357–368 These recent developments provide an exciting new potential to detect ischemia to various tissues before it is evident clinically.^369–376 Pulse oximetry requires the measurement of the pulsatile (arterial) component of the total light transmitted by the biologic tissue. Because NIRS does not require the subtraction of the pulsatile component, the signal is more than 100 times stronger and unaffected by poor perfusion compared with pulse oximetry. However, it is susceptible to motion artifact and ambient light noise but software algorithms have significantly decreased these sources of error. A probe placed on the child’s forehead measures the concentration of oxyhemoglobin and deoxyhemoglobin in the tissue underlying the probe. The device functionally measures the oxygenation of blood in the underlying tissues, including blood in the arterioles, capillaries, and venules. The majority (approximately 85%) of the signal originates from the venules; skin and bone and extracranial blood absorb only limited amounts of light, which is subtracted from the signal and does not have significant impact of the measurement in infants and children.^357,358 NIRS thus measures the O₂ saturation in venous and arterial blood and thus represents an average saturation across blood vessels and tissue. In cerebral oximetry, the O₂ saturation of the blood in the tissue (i.e., between the sending and emitting probe) depends on factors that affect O₂ transport, including cerebral blood flow, hemoglobin saturation, hemoglobin– O₂ binding affinity, and oxygenation saturation in the arterial blood. Therapies aimed at improving O₂ delivery or decreasing oxygenation consumption to the brain will potentially increase cerebral oxygenation (rSO₂).^359–361376

E-FIGURE 51-7 The pediatric clinical application of near-infrared spectroscopy relies on spatial resolution to extract surface tissue measurements, thereby isolating the target tissue, in this image of the neonatal cerebral cortex. The depth of measurement is resolved by the spacing of the light-emitting and light-receiving elements. Spatial resolution is illustrated in this diagram as arcs of red light traveling between light-emitters and light-receivers. Note that the short path measurements (A) correspond to intervening tissues, and long path measurements (B) correspond to the cerebral cortex. Light returned by the short path is deducted from total returned light, isolating the cerebral cortex.

(Courtesy Nonin Medical, Plymouth, Mont.)

This technology is not new; however, only three devices are currently FDA approved for commercial use to measure cerebral oxygenation: the first, is the INVOS Cerebral Oximeter (Somanetics Corporation, Troy, Mich.). The absolute accuracy of the current Somanetics device is ±10% to 15%, making the measurement of absolute oxygenation unreliable. However, it is reasonably accurate for changes in oxygenation (±5%), making this device most useful for following trends in cerebral oxygenation.^362–365 This device is currently FDA approved for trend monitoring. The Somanetics Cerebral Oximeter has demonstrated that normal regional cerebral oxygen saturation (rSO₂) is 60% to 80%.^366,367 In controlled hypoxic-ischemic states in animals, electroencephalogram (EEG) slowing and increased tissue lactic acid levels occur at rSO₂ between 40% and 45%. The EEG becomes flat at rSO₂ between 30% and 35%, and if the cerebral ischemia is protracted, it may be associated with tissue infarction.³⁶⁷

The second device, the Fore-Sight Absolute Tissue Oximeter (CAS Medical systems, Branford, Conn.) has been approved for actual saturation estimation and not just trending. This device uses four wavelengths of light and multiple sensors incorporated into one sensor, which is usually placed over the forehead; three sensor sizes are available, and use is based on patient weight.

The most recent device to be approved by the FDA is the EQUANOX (Nonin Medical Inc., Minneapolis, Minn.), which uses light-emitting diode (LED) technology to transmit three wavelengths (EQUANOX Classic Sensor) and most recently four wavelengths (EQUANOX Advance Sensor, Model 8004CA).^368,377 The EQUANOX sensor differs from the others by having dual emitter and sensor sites. The company claims this technology decreases surface tissue variability and can obtain quicker measurements. The first sensor did not have sufficient accuracy for absolute measurements and would be best used in trend monitoring. The newest sensor claims to be very accurate (4.1%), but further study needs to confirm the accuracy of these new systems.

This technology has important limitations. The relationship between the amount of tissue damage and clinical outcome depends on the site of the tissue damage. Because these monitors only measure the saturation in tissue below the probes, they more accurately reflect on focal tissue oxygenation or damage. If the probe is placed on the forehead, it will reflect the state of the tissue in the frontal cortex and may not reflect other areas of the brain.

Although some studies suggest improved outcomes when cerebral oximetry is used including pediatric congenital heart surgery, in which neurologic complications are high (2% to 25%),^{362,366,378,379} a fair degree of skepticism remains regarding the utility of cerebral oxygenation in clinical practice.^380,381 There is hope, however, that these devices may at some time in the future provide real-time guidance of regional tissue oxygenation and the effectiveness of interventions to improve the oxygenation. Additional research, further improvements in the technology, and reduced per-patient costs are needed to clarify the role of reflectance oximetry in pediatric patient management.³⁷³

Carbon Dioxide Analyzers

The most commonly used gas monitor is the expired CO₂ analyzer. This monitor is particularly useful in teaching about ventilation, CO₂ elimination from alveoli, and airway management, provided the response time of the instrument is rapid enough to accurately reflect end-expired CO₂ tension.^382,383 Two configurations of infrared CO₂ monitors are in clinical use. They differ only in the location of the CO₂ analysis: (1) the side-stream analyzer, which involves the aspiration of expired gas from the anesthesia circuit and remote CO₂ analysis, or (2) a mainstream analyzer in which an optical sensor within a cuvette is inserted into the circuit just proximal to the elbow and where analysis of the CO₂ occurs based on a signal detected at the cuvette. Side-stream analyzers are less cumbersome than mainstream analyzers because the former requires no additional pieces to be inserted into the anesthetic circuit. Gas for side-stream analysis only requires a narrow-gauge flexible tube to transfer the expired gas unidirectionally from the elbow of the circuit to the analyzer. The gas sampling line and water trap can become obstructed with water or secretions, requiring replacement of the parts. Mainstream gas analyzers do not obstruct with secretions, but they are heavy, and if a tracheal tube of a small caliber is used, airway obstruction may occur as a result of kinking of the tracheal tube.³⁸³ Furthermore, mainstream systems require a large inventory of cuvettes, which must be sterilized between patients. Most OR monitors today use side-stream sampling. The current generation of these monitors have improved precision that reduces the gas sampling rate to just 50 mL/min and excellent accuracy even with small rapid tidal volumes.^384–388

Capnography is the gold standard to ensure that the tracheal tube has been inserted within the trachea.^188,389 If no expired CO₂ is detected after a presumed tracheal intubation, one must assume that the endotracheal tube is in the esophagus. However, if the stomach has been distended by manual ventilation or crying before tracheal intubation or if carbonated soft drinks were present in the stomach, CO₂ could be detected immediately after intubation, even after an esophageal intubation.³⁹⁰ In such a case, the expired CO₂ will be very low or decrease rapidly in a stepwise manner after placing the tracheal tube. This artifact is very short lived. In addition, the presence of CO₂ during cardiopulmonary resuscitation is a very important sign of pulmonary blood flow.³⁹¹ Alternatively, its absence during resuscitation may indicate inadequate chest compressions and pulmonary blood flow (see Chapter 39).^392–395

Measurement of expired CO₂ tension is also helpful in detecting other clinical problems. Clinically important air embolism causes a transient but marked reduction in CO₂ excretion because the lungs are ventilated but not perfused; hence dead space is suddenly increased (see Chapter 24). Quantitative measurement allows detection of change in circuit flows, disconnections, endotracheal tube kinking, or accidental extubations.^{296,396–398} Capnography also signals the earliest clinical sign of a malignant hyperthermic reaction and the effectiveness of treatment³⁹⁹ (see also Fig. 40-3). In critically ill neonates or children in a hypermetabolic state after thermal injuries, ventilatory requirements may be as much as 4 times normal and capnography may be an effective noninvasive means of assessing the adequacy of ventilation in these children as well (see Chapter 34).⁴⁰⁰

A capnograph with a recorder or a display is most valuable because the CO₂ waveform may help in diagnosing other types of respiratory difficulties (Fig. 51-10). Bronchospasm and its response to treatment may be identified by changes in the CO₂ waveform; bronchospasm causes an increasing slope of the plateau during expiration, and resolution of the bronchospasm restores the plateau to the horizontal. Data also suggest that the shape of the CO₂ waveform changes with growth and development. The shape of the CO₂ waveform achieves the normal adult configuration by adolescence.⁴⁰¹ A slow-speed recorder also allows trending, which we have found particularly useful in diagnosing small circuit leaks, partially kinked endotracheal tubes, and rebreathing. In children free from pulmonary shunts, the arterial and alveolar CO₂ values (Paco₂ and PAco₂ [as reflected by a true end-expired sample]) should be within 2 to 3 mm Hg. The severity of a shunt or the diagnosis of a shunt may be made if this difference is greater than 5 mm Hg and if the CO₂ sensor is properly calibrated. Children with a variety of pulmonary problems may have significant differences between arterial and expired CO₂ values. In such cases, expired CO₂ monitoring may be used only for trending and as a disconnect alarm.^402,403 This monitor is of great value in assessing the adequacy of ventilation in children whose trachea is intubated, but of possibly less value in children whose airway is managed by facemask when it is difficult to obtain an accurate end-expired sample.^294,396 The accuracy of ETco₂ monitoring is slightly better with an LMA than with a facemask.⁴⁰⁴

FIGURE 51-10 Expired carbon dioxide (CO₂) tracings (A, B, C = rapid recording; D, E, F = trend recording). A, Normal waveform with a long alveolar plateau indicating good alveolar gas sampling during controlled ventilation. B, Spontaneous ventilation with rapid respiratory rate; minimal alveolar plateau. C, Patient with partial muscle paralysis; note change in CO₂ waveform (arrow) during inspiration, which took place during the ventilator expiration. D, Poor mask fit with many periods when no CO₂ was detected; this also results in many false alarms. E, A totally kinked endotracheal tube was detected by the absence of a CO₂ waveform (between arrows); a similar trace could result in a circuit disconnect, esophageal intubation, or a simple pause in respiration. F, A partially kinked endotracheal tube may result in a slow change in peak expired CO₂; a similar change could be noted with an unrecognized endobronchial intubation, change in pulmonary compliance, circuit leak, or increase in metabolic rate. The reverse would occur with air embolism, hypothermia (decreased metabolic rate), improved compliance, or, increased fresh gas flow without compensatory change in ventilator settings.

(From Coté CJ, Liu LM, Szyfelbein SK, et al. Intraoperative events diagnosed by expired carbon dioxide monitoring in children. Can Anaesth Soc J 1986;33:315-20.)

No monitoring device is perfect, and it is possible to develop a false sense of security when monitoring the ECG, O₂ saturation, and expired CO₂ values. Anesthesiologists must not abandon stethoscopes and the powers of observation; the practitioner is still the ultimate monitor. It is our powers of observation and clinical judgment combined with our ability to synthesize and integrate the data that determine the appropriate course of action. The following case is an example:

Although this is an extreme example, it emphasizes that even in the presence of highly sophisticated, properly working, noninvasive monitors, they may fail to accurately depict physiologic derangements in some clinical scenarios. Recognizing the limitations of our monitors and their potential for errors and remembering to verify physiologic indices with blood gas analyses (either arterial or venous) in the critically ill can only improve the quality of care delivered to the smallest and most critically ill infants and children under our care.

Anesthetic Agent Analyzers

Most current anesthetic agent analyzers use ultraviolet or infrared monochromatic (only one agent detected) or polychromatic (more than one agent detected) absorption or mass spectroscopy and provide a rapid time response.^418,419 Polychromatic infrared analyzers are most useful because they can detect erroneously selected anesthetic agents.⁴²⁰ Careful assessment of the end-expired vapor tensions may provide some indication of a child’s anesthetic depth, but each child must be assessed on an individual basis. Vaporizer malfunctions and miss-fills may also be diagnosed. Mass spectroscopy may be particularly valuable in providing an early warning of air embolism (after a child has been denitrogenated), because the sudden appearance of expired nitrogen is abnormal (entrainment of room air into the circuit and release of nitrogen from extremities with a tourniquet also may provide a source of nitrogen).⁴²¹ The purported advantages of less waste of anesthetic gases and a more rapid awakening are effects that have not been proved. However, the ability to use low flows offers the potential for significant cost savings in the form of less waste of anesthetic agent and less pollution of the atmosphere.⁴²² The report of accumulation of methane with closed circuit anesthesia and the inaccuracy of agent analysis (methane additive to the agent used) is worrisome but can be obviated with simple periodic flushing of the system.^423,424

Blood Loss Monitors

Close observation of the surgical field is the best single monitor of blood loss. A small-volume trap on the suction line before the major evacuation trap is particularly useful for small children. Routinely weighing surgical sponges on a dietary scale, assuming that 1 g of weight is equivalent to 1 mL of blood, is also helpful. Greatest accuracy is obtained by immediate weighing as the sponges come off the surgical field, thus minimizing evaporative losses. Point-of-care testing devices such as the HemoCue (HemoCue AB, Ängelholm, Sweden) and the I-STAT (Abbott Point of Care Inc., Princeton, N.J.) provide a rapid means for assessment of hemoglobin values^425–431 but each may underestimate the hematocrit at lower values.^{430,432–437} Noninvasive hemoglobin measurements using multiwavelength pulse oximetry in general may be useful for trend monitoring.^328,438 However, overall it tends to underestimate hemoglobin values and may not be sufficiently accurate at this time to use these data for clinical transfusion decisions because discrepancy of up to 1 g of hemoglobin have been reported.^{327,329,436,437,439–442}

Neuromuscular Transmission Monitors

Measuring the indirectly elicited twitch response is indicated when neuromuscular blocking agents are used. Residual or incomplete reversal of neuromuscular blockade has been shown in adults to be a common cause for delay in leaving the OR, prolonged postanesthesia care unit (PACU) recovery time, and respiratory complications in the early postoperative period.^443–446 It has been clearly determined that even experienced anesthesiologists cannot determine decrement in the train-of-four (TOF) accurately. Clinical signs (e.g., sustained head lift) are often difficult to perform in older patients and can be done with a significant degree of residual blockade, thus leading to undetected residual neuromuscular blockade and presumably the potential for respiratory complications.⁴⁴⁴ In general, two types of neuromuscular monitoring devices are available, those using acceleromyography and those using mechanomyography. Mechanomyography models use constant current nerve stimulation. These devices self-calibrate and then deliver a constant stimulus (30 to 70 mA) to the skin regardless of other influences. Failure to deliver a supramaximal stimulus may result in overestimation of twitch suppression. The power should generally be left off between readings to avoid possible injury secondary to prolonged, repetitious exposure to electric current. Acceleromyography uses a piezoelectric sensor to quantitate the acceleration (proportional to force) of the thumb converting this to an electrical signal. There is considerable disagreement in the published literature as to which of these techniques is most accurate.^447–449 Monitors based on acceleromyography are becoming more commonly available; however, these are difficult to use in infants because of the small arc of the displaced thumb. Some consider this monitor to be more accurate than the standard mechanomyography based train-of-four monitors, but they are difficult to use in infants and are not user friendly.^445,446 Others think mechanomyography is more accurate because it is less influenced by external disturbances, that is, it does not go out of calibration.⁴⁵⁰ Therefore at present for clinical purposes in infants and children, mechanomyography still seems to be the simplest and most helpful clinical monitor during the use of nondepolarizing competitive blockers. The TOF does not require recording capability or baseline measurement because it works strictly by comparing the fourth twitch with its own internal standard, the first twitch.^451,452 The unit may also be used with depolarizing relaxants, monitoring only the first twitch amplitude. A significant decrement (>50%) between the first and fourth twitch in this context suggests phase-two blockade (see E-Fig. 6-14).⁴⁵³ Needle electrodes are hazardous because they may cause bleeding, infection, burns, or nerve injury; current density may be very high, owing to the limited contact surface and the resistance of surrounding skin. Needle electrodes should not be used. Self-adhesive electrodes designed specifically for twitch monitors are available, but they are unnecessary and expensive and tend to be too large for infants and small children. A reasonable alternative is a pair of infant-sized ECG electrodes. The choice of monitoring location depends on the nature of the surgery. Any site where a motor nerve is close to the body surface and its associated muscle group is available for observation may be chosen. The most common stimulation site is the ulnar nerve in the forearm, observing the thumb (adductor pollicis brevis); this is the standard site for most research reports. It should be noted that the majority of studies have been standardized to the ulnar nerve, thus quantitating the response of other motor nerves is less reliable. Using the facial nerve may result in false-positive responses because the muscles may be easily stimulated directly, potentially causing an overdosing of nondepolarizing neuromuscular blocking agents.

Processed electroencephalogram Monitors

Monitoring the processed EEG has become a common means for monitoring patients who are anesthetized or sedated, but its role has yet to be fully defined in the pediatric patient population.^454–457 Analysis of the EEG waveforms and EEG data from brain response to auditory input have been used to assess the depth of sedation and anesthesia and in turn to predict or avoid awareness under anesthesia in adults. Several excellent reviews are available regarding the technology and process of validation.^458,459 Awareness under anesthesia has become a major source of litigation in the United States,⁴⁶⁰ and public concern prompted The Joint Commission to issue a Sentinel Event Alert.⁴⁶¹ In response to this alert, the ASA instructed a taskforce to review the available literature and the existing monitors.⁴⁶² Their review included studies from several processed EEG monitors (Bispectral Index [BIS], Aspect Medical Systems, Natick Mass.; Entropy, General Electric Healthcare Technologies, Waukesha, Wis.; and Narcotrend, MonitorTechnik, Bad Bramstedt, Germany) and data for auditory evoked potentials (AEP Monitor/2 (Danmeter Aps, Odense, Denmark). This task force reached several important conclusions:

A few studies have suggested that the use of these monitors may decrease awareness in adults^463,464 but no definitive data exist in infants and children that these monitoring systems decrease the incidence of awareness under anesthesia.⁴⁶⁵ The vast majority of studies have examined the Bispectral Index (BIS); therefore this section will primarily focus on this device with the understanding that other devices are available but have not been adequately evaluated in children.

Bispectral analysis generates a value (0 to 100) derived from processed electroencephalographic data that provide a measure of the depth of anesthesia.^458,466,467 There have been significant inconsistencies in data obtained from the BIS monitoring system in adults and even more inconsistencies in children. Some of the complexity and inconsistency of the BIS data will be reviewed to understand the possible uses and limitations of this and other EEG-based systems. Because of the nature of the data collection in the awake patient, artifacts may occur as a result of eye movement, blinking, and so forth at baseline if the device is applied before induction. Depth of anesthesia as measured by the bispectral analysis is more reliably assessed during propofol or inhalation-based anesthesia than during opioid-based anesthesia, although there is significant individual variation in the bispectral analysis reading.^468–470 N₂O has a minimal effect on the bispectral analysis, whereas ketamine paradoxically increases the bispectral analysis.^{458,471–473} Some well-controlled studies in adults examined the effects of a broad spectrum of medications on the bispectral analysis reading; this device monitored the effects of anesthetic drugs on awareness and patient responsiveness to painful and nonpainful stimuli.^{463,464,468,469,474–479} Studies also have demonstrated decreased anesthetic requirement and improved recovery parameters in adults who were monitored with bispectral analysis.^480,481 However, although bispectral analysis technology appears to monitor the effects of hypnotics, it does not necessarily monitor the adequacy of anesthesia or analgesia.⁴⁶²

Bispectral analysis values in children anesthetized with sevoflurane were inversely proportional to end-tidal sevoflurane concentration in both infants and children, although the BIS values in anesthetized children had a much broader range than in adults.^482,483 In one study, the BIS values decreased with increasing end-tidal concentration of sevoflurane until 3%, after which the bispectral analysis value paradoxically increased with higher inspired concentrations.^482,483 At equi-minimal alveolar concentration (equi-MAC), bispectral analysis values are greater in children than in adults.^483–485 Additionally, the bispectral analysis values at equi-MAC concentrations of volatile anesthetics has been inconsistent; several studies have shown that the bispectral analysis values during halothane anesthesia are greater than those during isoflurane, sevoflurane, or desflurane in children.^486–488 These data indicate that BIS values may be substantially greater with halothane than with the other inhalational anesthetics.⁴⁸⁹ In addition, several studies have demonstrated conflicting results in children compared with infants, suggesting greater bispectral analysis values for halothane in children but not in infants at equi-MAC.^482,485,488 All of these studies have suggested wide interindividual variability in the bispectral analysis values at comparable expired MAC concentrations of inhalational anesthetics. One disturbing study has found lower BIS values in intellectually impaired children compared with controls.⁴⁹⁰ Other studies have found differences between the right and left side of the brain.⁴⁹¹ Limited data validate bispectral analysis readings in young children and infants. The algorithm was based on processed EEG readings from adults, and the EEG does change with age developmentally; therefore it seems reasonable that the algorithm will need to be adjusted for age if the bispectral analysis is to be used in infancy.

One study noted an incidence of awareness in children aged 5 to 12 years of approximately 0.8%.⁴⁹² This is fourfold to eightfold greater than the incidence of awareness reported in adults, although the clinical implications are not clear at this time. Indeed, unlike in adults, none of the children who had awareness reported distress as a result.^493,494 The use of N₂O (when there is no contraindication) may be valuable in reducing the incidence of awareness without significantly adding to the duration of emergence.⁴⁹⁵ No randomized, controlled trials in children have demonstrated a decrease in awareness using bispectral analysis or other commercially available processed EEG devices. A recently introduced device is the SEDLine (Masimo Corp., Irvine, Calif.). This device uses four EEG electrodes and therefore simultaneously measures both sides of the brain. Preliminary experience at one institution has found that N₂O interferes with SEDLine monitoring (personal communication, CJC). At present, no validating studies in children have been reported; until further data are developed, we do not recommend adopting this technology.

The BIS has the longest track record, but again its value in children is still quite unsubstantiated. One study in children 3 to 18 years of age demonstrated a more rapid recovery when bispectral analysis monitoring was used. However, this did not hold true for children younger than 3 years of age.⁴⁹⁶ Bispectral analysis monitoring is able to help predict voluntary patient movement during intraoperative wake-up tests during spinal fusions.⁴⁹⁷ Bispectral analysis also has been used to assess the depth of sedation in children and to correlate bispectral analysis values with other measures of recovery, exclusive of ketamine-induced sedation.^498–500 Unfortunately, bispectral analysis values were unable to predictably separate moderate and deep sedation in children. In yet another study, a poor correlation was reported between the bispectral analysis values and a standardized Ramsay sedation score.⁵⁰¹ Bispectral analysis values do not appear to correlate with arousal from sevoflurane anesthesia.⁵⁰² Limitations on the usefulness of bispectral analysis monitoring in children include fear of probe placement while they are awake (thus in most cases the bispectral analysis monitor is applied after anesthetic induction). Because many procedures in children involve the head (e.g., strabismus, tonsillectomy, neurosurgical), the use of this monitor may be impractical in nearly half of the surgeries that involve children.

At this point, bispectral analysis and other depth of anesthesia monitoring systems appear to have some potentially useful applications in older children, whose EEG pattern is likely to be similar to that of adults, such as in trauma and during certain surgeries in which anesthetic drug use may be extremely limited (i.e., motor evoked potential monitoring for scoliosis surgery). However, the variability of response in any given child, the issues relating to the effects of specific drugs, and other technical factors such as probe placement in awake children, create significant limitations to its use in infants and toddlers. It is hoped that further, controlled studies will better define the possible role of depth of anesthesia monitoring in infants and children during sedation and general anesthesia.

Intracranial Pressure Monitors

In children with increased intracranial pressure (ICP), direct monitoring of ICP may be helpful. Measurement may be accomplished by transduced pressures through either a ventricular cannula placed through a burr hole or by a subarachnoid bolt or equivalent. Measurement also can be performed noninvasively in infants with an open anterior fontanelle (see Chapter 24).

Disconnect and Apnea Alarms

When a ventilator is used, the tactile feedback from palpating the breathing bag is lost. A disconnect alarm is useful in this circumstance, particularly in the absence of capnography. Most anesthesia ventilators incorporate a disconnect alarm, but it may also be purchased as a separate add-on for older designs. These alarms generally sense time cycling of pressure events. Most of these alarms can detect complete ventilator disconnects but may not be sensitive enough to detect extubations with small endotracheal tubes or partial circuit disconnects because of the continued presence of flow resistance. Other types of alarms, such as heat-sensing devices or capnographs, provide a more reliable means of detecting inadequate ventilation, accidental extubation, or disconnect. Besides the sensing of pressure cycles, many newer monitors include high-pressure, continuous positive-pressure, and negative-pressure alarms. Some products include an automatic switch-on when shifting from hand to machine ventilation, thereby avoiding failure to monitor.

Apnea Monitors

Apnea monitors, whether based on transthoracic impedance, motion, or other patient parameters, are helpful in the perioperative and recovery room phases for former preterm infants younger than 60 weeks postconceptual age.⁵⁰³ Infants with a history of apnea spells and children with ongoing apnea spells are much more likely to develop apnea in the postoperative period. Even if a child has had a period of some months with a normal respiratory pattern at home, the anesthetic state may bring about a temporary return of apnea spells, and such a monitor should be used according to current guidelines (see Chapters 2, 4, 35, and 36).

Volume Meter and Spirometer

In some instances, monitoring a child’s respiratory gas flow may be desirable. A Wright turbine respirometer, Dräger positive displacement spirometer, and newer designs based on the pneumotachograph operating principle, vortex detectors, or thermistor flow meters are available. Some devices allow measurement of flow–volume loops, which can be very useful in both diagnosing and correcting ventilatory issues related to bronchospasm, airway compression, or single-lung ventilation.^504–509 These monitors are not widely used in the clinical anesthesia care of children.

Inspiratory Pressure Gauge

In children who require ventilatory assistance, a pressure gauge applied at the airway allows the delivery of precise airway pressures and thus helps minimize barotrauma. When infants are treated with PEEP and continuous positive-airway pressure, airway pressures are best measured as close to the airway as possible. Such a device is useful in accurately determining leaks around an endotracheal tube when used to determine the size of the larynx (see Chapter 31).⁵¹⁰

Urinary Catheter

Aseptically inserted, a catheter in the bladder connected to a closed drainage system is of little risk to most children. Urine output is a useful indicator of volume and perfusion status; the presence of hematuria or hemoglobinuria can aid in the diagnosis of surgical trauma to the urinary system, a bleeding dyscrasia, or a transfusion reaction. Indications for a urinary catheter include massive fluid shifts, prolonged radiologic procedures with large doses of contrast material, prolonged surgery, neurosurgery and cardiac surgery with osmotic diuretic therapy, and urinary reconstructive procedures. In infants, standard urinary catheters might be difficult to insert and therefore a feeding tube provides an excellent substitute. Because the urine output in infants may be scant, connecting a feeding tube to a syringe barrel may provide an adequate and accurate measurement system. Care must be taken to avoid using latex catheters, particularly in children who are latex sensitive.

Invasive Blood Pressure Monitors

When clinically indicated, arterial, central venous, or very rarely pulmonary artery pressure should be monitored. Percutaneous arterial cannulation is performed with a 22- to 24-gauge catheter-over-needle device in infants and neonates, respectively, and a 20-gauge device in older children (see Chapter 48 and Fig. 48-10, A to D).^511,512 Central venous and pulmonary artery cannulation are usually achieved using the Seldinger technique (see also Fig. 48-2, A to D).^513,514 The proper reference location for all pressure transducers is usually the level of the child’s right atrium. If ICP monitoring is used or if the child is in the head-up or sitting position, a reference zero position at a reproducible cranial site (e.g., the external ear canal) is vitally important to accurately assess cerebral perfusion pressure. Placement of any of these invasive monitors provides additional benefits other than direct hemodynamic monitoring. Serial pH and arterial blood gas tension measurements from an arterial line or from determination of serial chemistries, ionized calcium, glucose concentrations, hematocrit, coagulation profiles, and osmolality allows continual assessment of a complex case. In operations performed in the sitting position, a right atrial catheter provides an approach to therapy for air embolism (see Chapter 24).

To withdraw a valid blood sample, an adequate volume of blood must be aspirated; 2 to 3 times the dead space (the point of aspiration to the tip of the catheter) usually accomplishes this.⁵¹⁵ In infants, these systems must be flushed with small volumes at slow rates. In infant arterial lines, this amount is 0.5 mL/5 sec, to avoid dangerous fluctuations in blood pressure and retrograde flow with the risk of stroke.⁵¹⁶

Echocardiography

Some reports have examined the usefulness of transesophageal echocardiography as a means of continuously monitoring cardiac output and the cardiac response to surgery and anesthetic agents. This equipment is very expensive and requires a great deal of skill and experience to take full advantage of its capacity. The practicality of this device in children appears to be primarily limited to those undergoing corrective cardiac surgery (see Chapters 14 to 16).^517–524 However, this device is clearly of benefit in defining the source of an adverse cardiovascular event such as an air, clot, tumor embolism, or other source of right-sided outflow tract obstruction.^525–529

Continuous Invasive and Noninvasive Cardiac Output Monitors

Traditionally, cardiac output has been estimated with thermodilution, Fick O₂ consumption, or echocardiography. Continuous invasive techniques such as continuous analysis of central venous O₂ saturation is feasible and accurate even in neonates but generally not applicable in the operating room.^530,531 Arterial pulse contour analysis is another invasive method that requires arterial and central venous catheters; data regarding the accuracy and reliability for children are still preliminary.^532–534 Other noninvasive techniques such as measurement of expired CO₂ have been attempted but with variable success.^535–542 The NICO device (Philips Respironics, Amsterdam, the Netherlands) estimates cardiac output as reflected by changes in expired CO₂ using a Fick partial rebreathing calculation.^541,542 This device requires that the child is intubated, and its accuracy in children is yet unproved.^{536,539,541,543} A new device, Cardiotronic EC (Cardiotronic, Inc., La Jolla, Calif.), has been approved by the FDA for use in children of all ages, including neonates.^544–549 Electrical cardiometry continuously estimates stroke volume and left ventricular outflow to calculate cardiac output, cardiac index, and a variety of other parameters through quantitation of changes in impedance associated with changes in the orientation of red blood cells (Fig. 51-11). During diastole red cells are organized chaotically, but during systole they assume a position parallel to the direction of blood flow. Thus thoracic electrical bioimpedance relates to changes in thoracic aortic blood flow, and through the use of refined algorithms, noninvasive measurement of cardiac output is achieved.

FIGURE 51-11 A, Cardiotronic cardiac output device. B, Electrical cardiometry continuously estimates stroke volume and left ventricular outflow to calculate cardiac output, cardiac index, and a variety of other parameters through quantitation of changes in impedance associated with changes in the orientation of red blood cells. During diastole red cells are organized chaotically, but during systole they assume a position parallel to the direction of blood flow. Thus thoracic electrical bioimpedance relates to changes in thoracic aortic blood flow and through the use of refined algorithms noninvasive measurement of cardiac output with four electrocardiography pads is achieved.

The height (cm), gender, weight (kg), and age of the patient are entered into the handheld device. Four electrodes are applied to the neck and left chest at specified locations—in infants, one lead is placed on the cheek or forehead and the fourth lead placed on a thigh (see Fig. 51-11). The device records the heart rate and averages cardiac output every 10 to 60 heartbeats depending on how the device is configured. Studies in neonates, children, and adults suggest that this is a reasonable trend monitor to estimate cardiac output in a noninvasive manner. It certainly is not a monitor that will replace the gold standard of echocardiography or thermodilution cardiac output, but the r² value of 0.86 and Bland-Altman plot suggest reasonable efficacy for monitoring and trending purposes in adults.⁵⁵⁰ Preliminary experience (CJC) with this device is favorable as an early warning device of a developing adverse cardiovascular event and as an indicator of successful intervention of such events.

Electrocautery

Although anesthesiologists are not directly responsible for electrocautery devices, they should have a basic knowledge of how they function and the necessary precautions to prevent burn injury or a fire. The most important means of preventing injury is to ensure that an appropriate-size grounding pad has been applied and that no blood, urine, or preparation solutions have been allowed to pool around or under the grounding pad. Additionally, the electrocautery can ignite any flammable implements, drapes, and an O₂-rich environment. Severe injuries and deaths have occurred when high concentrations of O₂ have been allowed to build up beneath drapes (supplemental O₂ in awake or sedated patients) and have been ignited by the spark of the electrocautery on the other side of the surgical drape.^551–556