Historical Development

In North America, improvements to trauma care began in the 1960s and 1970s (Trunkey, 1983; Haller, 1995; Mullins, 1999; Morrison et al., 2002). Pediatric trauma systems did not develop in isolation but in concert with adult care. Trauma care was advanced in both the Korean and Vietnam wars, with the military experiences of treating injured soldiers. A report by Howard (1966) highlighted the enormity of the injury problem and stressed its significance as a neglected public health problem. Furthermore, a series of “preventable death studies” was published that provided fodder for community and political support of coordinated care for injured patients. Funding for emergency medical services first became available in 1966 with the National Highway Safety Act, and further support followed with the Emergency Medical Services System Act in 1973 (Mullins, 1999). In 1968, Cook County hospital in Chicago (for adult care) and Kings County hospital in Brooklyn (for pediatrics) were recognized as the first specialized centers for civilian trauma care in the United States. In 1969, the University of Maryland and the Maryland State police developed a coordinated transport system for injured patients to preferentially take them to a hospital with specific interest in caring for injured patients (Cowley et al., 1973; Cowley and Scanlan, 1979). Illinois is credited as the first state to develop a comprehensive state-wide trauma system that included categorization of trauma hospitals and the establishment of a communication system and a trauma registry (Boyd et al., 1973). In 1984, the Department of Health and Human Services, in conjunction with the National Highway Traffic Safety Administration, began the Emergency Medical Services for Children Program in the United States. This funding has been a critical resource for developing programs and research to improve emergency service systems for children.

The American College of Surgeons Committee on Trauma (ACSCOT) has also strived to improve trauma care in North America. This has been accomplished through the publication of standards for trauma systems, the creation of educational courses for health care professionals, and the establishment of and verification programs to ensure published standards are met. In 1976, the first version of the Optimal Resources for Care of Seriously Injured was released (Committee on Trauma: American College of Surgeons, 1999). The most recent version of this publication (2006) is currently in use and has been retitled “Resources for the Optimal Care of Injured Patient.” It remains a dynamic and important document. This publication established criteria for defining levels of trauma centers (I through III) and trauma systems (I and II), and it defined resource requirements for prehospital care through discharge, as well as specific pediatric needs. Level I centers offer the widest range of services for the most severely injured patient, whereas level III centers allow for stabilization and triage. With respect to trauma system development, eight criteria were defined and have become accepted benchmarks for trauma care (Boxes 30-1 and 30-2). Recently, designation of a level IV trauma center has occurred in some states where the resources do not exist for a level III trauma center. Level IV trauma centers are able to provide initial evaluation, stabilization, preliminary diagnoses, and transportation to a center with a higher level of care. These centers may also provide surgery and critical care services (as defined in the scope of services of trauma care). A trained trauma nurse and physicians are available on the patients’ arrival to the emergency department.

In 1986, the ACSCOT established its consultation and verification program to ensure that these standards were being met. At present in North America, there are 94 verified level I centers for adult care and 46 level I centers for pediatrics.

Another valuable contribution by the ACSCOT is the Advanced Trauma Life Support course (ATLS). First released in 1979, the course is now taught all over the world (American College of Surgeons Committee on Trauma, 1997). The first pediatric chapter was introduced in 1983. Although initially designed for surgeons, a wide variety of health care professionals involved in trauma care participate today. The ATLS course provides a common language, framework, and approach to injured patients that facilitate communication between health care professionals in order to optimize and prioritize care. Regular recertification is mandated to keep physicians current with new advances in trauma management. Similar to other ACSCOT programs, the ATLS course is constantly evolving. The current version is the seventh edition.

Trauma System Components

A trauma system is not simply an isolated hospital that cares for injured patients but a broad coalition of participants that includes an integrated approach to trauma. This includes prevention, prehospital care, and hospital care through rehabilitation (Ehrlich et al., 2001, 2002).

A primary step in developing an effective trauma system is a needs assessment of the community that it serves. For example, rural trauma systems differ from those in urban settings—each have different mortality rates, injury patterns, geographic areas, available resources, and expertise (Rogers et al., 1999; Ehrlich et al., 2004). Community responsibility and involvement (particularly with injury prevention) are essential for political support and funding. Studies support the concept that community-based injury-prevention programs play a substantial role in reducing morbidity and mortality (Hulka et al., 1997; Nathens et al., 2000).

A second important consideration in developing a regional trauma system is to understand that all facilities have a role in treating injured patients. A tiered approach (e.g., Levels I through IV) with specific hospitals designated to care for the varied complexity of injured patients is essential. Level I hospitals are tertiary or quaternary centers, with levels II and III centers having fewer capabilities and resources. Level IV hospitals are distinct because of their remote nature (rural trauma). A list of recommended requirements for each level of care can be found in the ACSCOT publication Resources for the Optimal Care of the Injured patient: 2006 (Committee on Trauma: American College of Surgeons, 1999). A basic tenet and overarching requirement for this tiered system is a central communication (central command) structure and a well-developed triage system so that the most severely injured patients go first to the appropriately designated hospital. In addition, a trauma registry system is essential for outcomes analysis and quality-assurance programs.

A third point is that a trauma system is not an isolated surgical program but a multidisciplinary team of physicians, nurses, technicians, and allied health professions. A large trauma program uses a significant percentage of hospital resources for everything from the blood bank to social work. Regular meetings of all stakeholders are needed to solve problems and respond to changing injury patterns and community needs.

Anesthesiologists play a variety of roles in trauma response. These roles span from assisting as primary responders in airway management and vascular access to coordinating operating-room resources to ensure that patients are treated in a timely and effective fashion. A pediatric anesthesiologist is an invaluable resource when treating injured children. A 1993 report from the Institute of Medicine identified specific pediatric emergency medical service deficiencies in rural states (Institute of Medicine, 1993), including the pediatric airway management, vascular access, and special resuscitation needs of injured children. Airway control is a primary tenet of trauma care, and proper endotracheal intubation (ETI) is recognized as a definitive method to achieve airway control. ETI, however, in injured children is required less often than in adults; hence, maintaining skills can be difficult for health care providers (Baker et al., 2009). Several studies have looked at airway management in pediatric trauma and examined significant aspects, including how success and complication rates of ETI vary by location (e.g., field, transferring hospital, or trauma center) and personnel, whether bag-mask ventilation is effective with respect to oxygenation and ventilation before ETI, and whether multiple attempts at intubation adversely affect patient care or outcomes (Cooper et al., 1992; Gausche et al., 2000; Glaeser, 2000; Ehrlich et al., 2004). These studies have demonstrated that field intubation success rates are poor. Successful pediatric intubations relate to the skill of the person performing the intubation, and anesthesiologists’ airway management skills are considered superior to those of all other health care professionals. Failed attempts at ETI appear to produce a spiral effect that results in multiple failed attempts and delayed transfer to definitive care. Complications resulting from ETI seem far more common when the intubation is attempted in the field and the risk of complication is exacerbated with each attempt. Increased complication rates do impact patient recovery, as shown by longer durations of hospitalization (Cooper et al., 1992; Gausche et al., 2000; Glaeser, 2000).

Triage Scores: Where Is the Injured Child Best Treated?

An ideal triage scoring system would have few data points, be easy to apply, have limited subjective assessments, and have a high sensitivity and specificity. Unfortunately, no triage score accomplishes all of these tasks, but each system does have strengths and limitations. For example, studies have shown that assessments made in the field by paramedics are less accurate and reliable for pediatric trauma patients when compared with adults (Engum, 2000). Alternatively, comparisons between adult and pediatric trauma triage scores have not shown that a pediatric specific trauma score provides a significant advantage over the original adult measure.

Glasgow Coma Scale

In 1974, Teasdale and Jennett first introduced the GCS. A lower score reflects a lower level of consciousness and therefore a potentially more serious head injury. Scores can range from 3 to a maximum of 15 (normal). The GCS measures three specific components of consciousness (eye movement, verbal, and motor responses) with higher scores (up to 5) given to the best response. A patient with a GCS of 13 to 14 is considered to have a mild head injury; 9 to 12 indicates a moderate injury; and a score of 8 or lower indicates a severe insult. Adaptations of the GCS for the pediatric population have occurred, and a Pediatric GCS is now widely used (Table 30-1).

GCS scores determined in the field are less predictive of outcome than those generated at a hospital (Meredith et al., 1995). When each individual component of the GCS is evaluated, the motor component is the strongest predictor of outcome.

Revised Trauma Score

The revised trauma score (RTS) is a physiologic-based triage score (Table 30-2). The RTS was derived from two earlier versions of a triage scores, the Triage Index and the Trauma Score (Champion et al., 1980, 1981, 1989). The RTS has three variables—respiratory rate, systolic blood pressure, and GCS. The RTS is the sum of each variable multiplied by a weighted coefficient.

TABLE 30-2 Revised Trauma Score*

* The Revised Trauma score is the sum of the weighted variables (see formula in text). The higher the score, the better the prognosis.

From Centers for Disease Control and Prevention: Traumatic brain injury in the United States: a report to Congress, Atlanta, 1999, Department of Health and Human Services National Center for Injury Control and Prevention.

where GCS is the Glasgow Coma Scale; SBP is systolic blood pressure; and RR is respiratory rate. These variables were determined to correlate statistically with survival and mortality. A higher RTS is associated with a better chance of survival. An injured patient with a RTS score of 11 or lower is recommended to be treated at a designated trauma center. Eichelberger established and developed pediatric coefficients for the RTS that was then validated in the pediatric population (Engum et al., 2000). The RTS is also recommended as one of the triage tools in the American College of Surgeons trauma guidelines, Resources for the Optimal Care of the Injured Patient (Committee on Trauma: American College of Surgeons, 1999).

Alert, Verbal, Painful, or Unresponsive Scale

This is a simple scale that categorizes a patient’s condition as alert, verbal, painful, or unresponsive (APVU). No numeric scores are given, but an injured child with a P or U should be taken to the designated trauma center. Subjective interpretation is inherent in this score, so it is best used as a component of the triage process, especially for children in whom the age and development of the child can impact the response. There is a significant risk of undertriaging patients when using this score.

Pediatric Trauma Score

The Pediatric Trauma Score (PTS) was first introduced by Tepas in 1987 (Tepas et al., 1987; Nayduch et al., 1991). Its design was thought to be more reflective of pediatric injuries (Table 30-3). Six variables comprise both physiologic and injury attributes. After a number for each variable is totaled, a score of 8 or lower is considered to be a marker for children requiring care at a designated trauma center. A limitation of the PTS is that some variables are subjectively scored, particularly the central nervous system (CNS) assessment. Therefore, there is a risk of interrater and intrarater variability. The PTS has been compared with the RTS with surprisingly few apparent differences or inherent advantages (Tepas et al., 1987, 1988; Nayduch et al., 1991).

TABLE 30-3 Pediatric Trauma Score

* A total score of 8 or less suggests care at a designated trauma center.

Data from Tepas JJ III, Mollitt DL, Talbert JL, et al: The pediatric trauma score as a predictor of injury severity in the injured child, J Pediatr Surg 22:14–18, 1987; Tobias JD, Ross AK: Intraosseous infusions: a review for the anesthesiologist with a focus on pediatric use, Anesth Analg 110:391–401, 2010.

Age-Specific PTS

The most recent proposed triage score is the Age Specific PTS (Table 30-4). First described by Potoka et al. in 2001, the main addition is the use of age-specific physiologic variables. To date, long-term data are lacking and only a single study suggests that it is better than traditional adult scores; therefore, the clinical utility as a triage score still remains to be proven. This score is, however, a pediatric-specific tool that addresses areas where other adult scores become less accurate (e.g., extremes of age).

Abbreviated Injury Scale

The Abbreviated Injury Scale (AIS) emerged from the automotive industry in 1969 (JAMA, 1971; Association for the Advancement of Automotive Medicine, 1990). It was initially designed as an epidemiologic tool to describe motorvehicle crashes but has been adapted to all types of trauma (JAMA, 1971; Association for the Advancement of Automotive Medicine, 1990). Revised and updated several times over the years, the first version published was the AIS-90 (for the year 1990) in 1998 (Association for the Advancement of Automotive Medicine, 1990). This score evaluates nine body regions. A scale from 1 to 6 is used to define injuries (1, minor; 2, moderate; 3, serious; 4, severe; 5, critical; and 6, maximal.) The underlying premise is an association of injury with threat to life. A body region with a score of 6 is considered an unsurvivable injury. The scores are determined retrospectively and based primarily on ICD-9-CM codes. A significant amount of expertise is required to assign these ratings, and interrater and intrarater variability is common. To help limit this phenomenon, software has been developed to convert the ICD-9-CM codes directly to an AIS score. Unfortunately, ICD-9-CM coding itself is inherently variable (MacKenzie et al., 1985, 1989).

Injury Severity Score

The Injury Severity Score (ISS) is one of the most widely used scoring systems in the trauma literature. It correlates well with several important trauma outcomes such as mortality and duration of hospitalization. Developed in 1974, the ISS is a method of characterizing the trauma patient with multiple injuries (Baker et al., 1974; Baker and O’Neill, 1976). The ISS is based on the AIS, which describes the severity of injury to different body regions (see the previous section). For example:

The above formula shows AIS scores of the three most injured of the following body regions: A, the head, neck, and face; B, the thorax and abdomen, and C, the extremities (including the external pelvis). The ISS takes scores from 0 to 75 (i.e., AIS scores of 5 for each category). However, if any of the three scores is a 6, the score is automatically set at 75, because a score of 6 indicates that an injury is unsurvivable.

Two modifications have been reported to the ISS score—the modified ISS (MISS) incorporating the GCS, and the New ISS, focusing on injuries over body regions. However, evaluations of both the MISS and new ISS have not shown advantages over the standard ISS; therefore, in many centers and registries the AIS-90 still remains the standard severity of injury scoring system.

Survival Probability: Trauma Score and Injury Severity Score

The Trauma Score and Injury Severity Score (TRISS) is not a score but rather a method of predicting mortality. It was first proposed in 1987 and combines the physiologic variables of the RTS with the anatomic severity-of-injury scores generated by the ISS (Boyd et al., 1987). The final result is that a probability of survival (Ps) is generated.

Where e is the base of the natural log 2.72183, and b is derived from the following formula:

The age factor is 0 for those younger than 55 years of age, and 1 for those who are 55 years of age. The variables b0, b1, b2, and b3 are blunt and penetrating trauma coefficients for adult and pediatric populations (Eichelberger et al., 1993).

TRISS is the most widely used predictor of survival in the trauma literature. As with other statistical processes, limitations exist. These are specifically noted in patients older than 55 years of age (extremes of age), and with trauma patients whose ISS is greater than 25. In fact, some authors suggest that because of the TRISS deficiencies, this method of calculating survival probabilities should be abandoned completely, particularly in urban centers (Cayten et al., 1991; Demetriades et al., 1998).

A Severity Characterization of Trauma Score

A Severity Characterization of Trauma (ASCOT) score is an attempt to revise and fix the limitations found using TRISS methodology (i.e., severely injured patients and extremes of age) (Champion et al., 1990b). ASCOT uses individual components of the RTS (not totaled) and incorporates a modification of AIS entitled The Anatomic Profile. Furthermore, it excludes those patients with either very severe or nonserious injuries (a maximum AIS score of less than 2 and an RTS score of greater than 0; an AIS score of 6 and an RTS score equal to 0). The Anatomic Profile categorizes the AIS scores that are greater than 2 into three groups: head, brain, and spinal cord injuries; thorax or neck injuries; and all other serious injuries. A comparison of ASCOT vs. TRISS with a pediatric population did not provide more reliable or accurate scoring over the TRISS.

Pediatric Age-Adjusted TRISS

This is a recent tool used to describe mortality that applies the pediatric-specific, age-adjusted PTS. Although it may be a more promising predictor of mortality, its data are limited and it has not been validated by other investigators (Schall et al., 2002).

The multitude of scoring systems reflect the reality that no single approach addresses all aspects of pediatric trauma accurately or provides the necessary tools to effectively answer all research questions. Many scoring systems suffer from inherent inaccuracies of coding (e.g., related to discharge diagnoses, ICD-9-CD diagnoses, and data on death certificates). Although physiologic variables generally seem to outperform anatomic variables in these systems, codification requires experience and training. Finally, certain elements of the laboratory data have been shown to be a strong predictor of mortality, yet these variables are not included in the prediction models.

Airway

The first step in trauma resuscitation is to assess the airway and then secure it when necessary to assure respirations. Indications for ETI include inability to oxygenate and ventilate and the need to protect the airway against aspiration. Appropriate management of the airway may be challenging or difficult without proper preparation and familiarity with the unique characteristics of the pediatric airway. In a young child or infant, the tongue is relatively large and the larynx and glottic opening are more anterior. A child’s airway has a smaller diameter and a shorter length. Airway edema occurring in an already small airway results in significant changes in the internal diameter of the airway and in increased resistance to airflow. The short length of the trachea makes right mainstem intubations more likely and increases the likelihood of extubation from small positional changes of the ET tube. Because the pediatric mid-face is small compared with the tongue, airway obstruction is common. Blood, vomit, teeth, and foreign bodies are common causes of airway obstruction. In addition, children are often placed in cervical collars that may be too large for them. This can make it hard see a child’s face and can present a challenge to identifying an airway obstruction. The first step to making the obstructed airway patent (while keeping the cervical collar on) is a jaw thrust with suction. Oral or nasal airways may be required if a jaw thrust is unsuccessful. If the airway is still not patent after these maneuvers, there should be consideration of a possible airway foreign body.

The initial management of the pediatric airway involves bag-valve-mask ventilation with a jaw-thrust maneuver. Intubation is indicated in patients with respiratory or cardiac compromise or an altered level of consciousness. All pediatric trauma patients should be considered to have a full stomach and possible cervical spine injury. See related video online at www.expertconsult.com.Because of this, the airway should be secured after a rapid-sequence induction (RSI) (that may include cricoid pressure) with manual inline stabilization and an oral ETI. Nasotracheal intubation may be suboptimal and difficult because of the small size and acute angle of the nasopharynx and the more anterior and cephalad position of the glottic opening.

The RSI for pediatric trauma patients can be accomplished with an induction agent that is immediately followed by a muscle relaxant. Standard induction agents for trauma patients include etomidate (0.2 to 0.3 mg/kg) and ketamine (2 to 4 mg/kg) or the combination of fentanyl (2 to 3 mcg/kg), midazolam (0.05 to 0.1 mg/kg) with lidocaine (1 mg/kg). Sodium thiopental (STP) and propofol should be reserved for patients who are not hemodynamically unstable. STP is an ideal induction agent for patients with head trauma, provided they are not hypovolemic. Ketamine is relatively contraindicated in patients with increased intracranial pressure (ICP). Although etomidate provides hemodynamic stability in trauma patients who are hypovolemic, it may decrease survival in patients with sepsis secondary to adrenal suppression (Annane et al., 2002). Muscle relaxation can be achieved with rocuronium (0.8 to 1.2 mg/kg) or succinylcholine (1 to 1.5 mg/kg). Succinylcholine is contraindicated in crush injuries, long-bone fractures, and patients susceptible to malignant hyperthermia. In patients who are hemodynamically stable, the combination of propofol (4 mg/kg) and remifentanil (3 mcg/kg) can be used for rapid-sequence intubation and has an onset and offset similar to propofol and succinylcholine (1 mg/kg) (Crawford et al., 2005).

ETIs have been deemphasized in the prehospital setting, because they are often unsuccessful. Gausche and others (2000) reported a success rate of only 57%. All intubated trauma patients that come to the emergency department should have the placement of their endotracheal tubes (ETTs) confirmed with either end-tidal carbon dioxide (CO₂) or direct laryngoscopy.

If initial intubation attempts are unsuccessful after an RSI, the patient should be ventilated with bag-mask ventilation. If a patient cannot be ventilated or if it is very difficult, a laryngeal mask airway (LMA) can be placed to facilitate ventilation and subsequent intubation. However, it must be recognized that the LMA does not protect the airway from aspiration and must be replaced by an ETT as soon as skilled personnel become available. The stomach should be decompressed with a nasogastric or orogastic tube after intubation, and a chest x-ray should be obtained to verify the ETT position (Fig. 30-2).

FIGURE 30-2 Gastric dilation often occurs after crying or positive-pressure ventilation by gas and mask.

Depending on the nature of the underlying injury, securing the airway in a patient who has sustained multiple injuries or even isolated facial injuries can be extremely complicated, as illustrated in Figure 30-3. The management of such cases calls upon the resourcefulness and skills of the anesthesiologist and requires careful consideration of damage to surrounding structures such as major blood vessels and the airway structures themselves. The ability to maintain a patent airway via face mask, and the potential for an expanding hematoma that may subsequently compromise an airway that may be patent at the current time, must be anticipated. Additional considerations include the risks of increased ICP with concomitant head trauma, exacerbating an existing cervical spine injury, and aspiration during airway manipulation. The presence of rhinorrhea, otorrhea, or ecchymoses around the eyes should raise suspicion about a possible basilar skull fracture, and any instrumentation of the nasal passages, including passage of a nasal ETT or an N/G tube, should be avoided. Similarly, crepitus at the neck may herald the presence of a tracheal disruption, and intubation under direct vision using a fiberoptic scope should be considered to avoid false passage of the ETT. See related video online at www.expertconsult.com.

FIGURE 30-3 An 11-year-old girl fell 10 to 15 feet while sliding down the school banisters onto a plant supported by a thick wooden pole. She eventually was intubated orally with direct laryngoscopy.

(From Melillo EP, et al: Difficult airway management of a child impaled through the neck, Paediatr Anaesth 11:615, 2001.)

In cases in which airway difficulty is anticipated, it may be prudent to transport the child to the operating room with an anesthesiologist and otolaryngologist once the child has been stabilized hemodynamically and additional injuries have been ruled out. The airway may then be secured with preparations to perform an emergent tracheostomy in case of failed laryngoscopy. An inhalational induction may be tolerated by the patient who has been volume resuscitated. This permits direct laryngoscopy or flexible fiberoptic intubation while the patient is breathing spontaneously. In patients who have suffered loss of consciousness or head injury, there should be a high index of suspicion for cervical spine injury, even with the absence of radiologic evidence. When performing laryngoscopy, inline axial stabilization must be performed (Fig. 30-4). Inline axial stabilization is performed by an assistant during laryngoscopy. The assistant should place both arms on either side of the patient’s head while gripping the patient’s shoulders. The assistant’s function is to maintain the patient in a neutral position during laryngoscopy, avoiding flexion, extension, or rotation of the cervical spine. The use of muscle relaxants is best avoided until the airway is secured. If intravenous agents are required to induce anesthesia, it is preferable to use short-acting agents such as propofol and remifentanil that effectively blunt ICP responses to direct laryngoscopy yet permit return of spontaneous respiration in case of failed intubation.

FIGURE 30-4 Manual axial inline stabilization during direct laryngoscopy.

Breathing

Assuring adequate ventilation is the next task after securing the airway. Breathing is best assessed by auscultation and observation of chest motion. During the primary assessment the patient should be assessed for the presence or absence of breathing, respiratory rate, and work of breathing. The chest should also be observed for symmetry of chest motion and breath sounds. The chest wall should also be observed for evidence of direct chest trauma resulting in abrasions, penetration, or chest-wall instability. The back needs to be assessed as well for posterior chest wall trauma. Although there are no concrete recommendations to guide intubation, pediatric patients with increased work of breathing or life-threatening injuries (head injuries) may require ETI. Immediately after ETT placement, the position should be confirmed with end-tidal CO₂. Patients in cardiac arrest or very small patients may require direct laryngoscopy. A portable chest x-ray and arterial blood gas should be obtained before leaving the trauma bay to determine the position of the ETT and its adequate delivery of oxygenation and ventilation. Continuous end-tidal CO₂ monitoring and pulse oximetry are useful adjuncts. Many injuries that impair respiration include simple, tension, and open pneumothorax; massive hemothorax; flail chest; and pulmonary contusion. A sudden change in the respiratory status after successful intubation should prompt an immediate evaluation for a reversible cause. An enlarged pneumothorax or development of a tension pneumothorax can occur after converting from spontaneous ventilation to positive pressure ventilation. Other causes for a decline in respiratory status include a displaced ETT (e.g., extubation or right mainstem intubation), obstruction in the ETT (e.g., blood or secretions), and equipment failure.

Circulation and Access

Children who sustain multiple injuries often arrive in hypovolemic or hemorrhagic shock that must be promptly recognized and treated. Unlike adults, children maintain an almost normal blood pressure until 25% to 35% of their circulating blood volume is lost (Fig. 30-5). This is likely because of their high sympathetic tone that causes peripheral vasoconstriction in an effort to maintain blood pressure in the face of a diminished blood volume. Therefore, tachycardia is an earlier sign of impending shock than hypotension. Additionally, signs of poor peripheral perfusion such as delayed capillary refill (more than 2 seconds), weak or thready pulses, mottling or cyanosis of the skin, and impaired consciousness are earlier indicators of shock than low blood pressure. The presence of hypotension as a result of hypovolemia should be considered an ominous sign that usually heralds impending cardiovascular collapse. Table 30-5 describes the stages of pediatric shock and clinical signs seen at these stages.

FIGURE 30-5 Increase in systemic vascular resistance in response to hypovolemia preserves blood pressure until 25% of blood volume is lost. Hypotension is a late sign of hypovolemia.

(From Rasmussen GE, Grandes CM: Blood, fluids, and electrolytes in the pediatric trauma patient, Int Anesthesiol Clin 32:79, 1994.)

TABLE 30-5 Stages of Pediatric Blood Volume Loss (Shock) and Associated Clinical Signs

CV, Cardiovascular; CNS, central nervous system.

From Rasmussen GE, Grandes CM: Blood, fluids, and electrolytes in the pediatric trauma patient, Int Anesthesiol Clin 32:79-101, 1994.

It is imperative to rapidly assess the pediatric trauma patient for signs of shock upon arrival in the trauma center and at regular intervals thereafter. The initial fluid bolus administered in the trauma setting is warmed isotonic crystalloid (lactated Ringer’s solution or normal saline) in a bolus of 20 mL/kg, IV (see Fig. 30-1). The pulse, capillary refill, and blood pressure are reassessed. A second bolus of 20 mL/kg is administered if there is no significant response or only a transient improvement in these parameters. A third crystalloid bolus may be given if necessary to maintain appropriate vital signs and circulation. Blood (10 mL/kg) should be administered next if additional fluid resuscitation is required. The need for blood transfusion initially is uncommon and usually signals surgical bleeding that may require an operation.

If shock persists and fails to respond to fluid therapy, other causes should be sought. Such causes may include long bone or pelvic fractures. Pericardial effusion and tamponade are less common occurrences in blunt trauma compared with penetrating injuries. The classic clinical signs of cardiac tamponade are shock, muffled heart sounds, pulses paradoxus, electrical alternans, and distended neck veins. Treatment requires immediate pericardiocentesis.

Pneumothorax is a common complication of blunt chest injury in children, with nearly one fourth of pneumothoraces under tension (Nakayama et al., 1992). Unilateral or bilateral tension pneumothoraces may produce hypotension and hypoxemia. The classic signs of tension pneumothorax are ipsilateral tympany, shift of the trachea to the contralateral side, and distended neck veins.

Significant occult blood loss may be overlooked on the initial examination of the small child and infant. Because the absolute blood volume of a child is small, the significance of external blood loss may be underestimated. In addition, blood accumulation in the infant’s large, expandable head and open fontanels can produce shock. Careful assessment of the abdomen is also central in the evaluation of the injured patient in shock.

Adequate large-bore IV access must be established as early as possible in the course of the resuscitation. Although peripheral routes offer the most rapidly accessible sites, such access may be difficult or impossible to obtain in the child with depleted intravascular volume or shock and resultant peripheral vasoconstriction. In such cases, a central venous catheter may be placed in the femoral or neck vessels if personnel with the necessary skills are available. Delays in establishing vascular access may be lifethreatening; therefore, if the above routes fail, intraosseous (IO) access should be rapidly initiated to expedite the administration of volume expanders and necessary pharmacologic agents. Once the child has been resuscitated, additional vascular access should be obtained.

IO access is placed in the medial surface of the proximal tibia 1 to 3 cm below the tibial tuberosity or the distal femoral metaphysis. IO has been used as a lifesaving measure to establish short-term vascular access in critically ill or injured children (Fig. 30-6).

FIGURE 30-6 Appropriate placement of the IO infusion needle on the medial surface, distal to tibial tuberosity.

(From Ellemunter H: Intraosseous lines in preterm and full-term neonates, Arch Dis Child Fetal Neonatal Ed 80:74F, 1999.)

The use of IO infusions in the algorithm of trauma and cardiopulmonary resuscitation has evolved over the years and now takes a more prominent role for providing vascular access to children. The subject of IO has been reviewed by Tobias and Ross (2010).

Access needles for IOs include manual devices such as butterfly needles, spinal needles, bone marrow biopsy needle such as Susfast (Cook Critical Care; Bloomington, Indiana) and the basic Jamshidi needle (Baxter Healthcare Corporation; Deerfield, Illinois). More recently, automated devices have been marketed (Fig. 30-7). The EZ-IO (Vidacare; San Antonio, Texas) functions like a battery-powered drill, with a beveled drill and a preset depth. It is designed to be used in the tibia. The Bone Injection Gun (BIG; Waismed; Kansas City, Missouri) is a spring-locked device also designed to be used in the tibia. It comes in a range of sizes for use in infants, small children, and adults. A device specifically designed for entry into the bone marrow of the sternum is the FAST1 (First Access for Shock and Trauma, PYN6 Medical Corporation; Vancouver, British Columbia). It is approved for patients ages 12 and older. Table 30-6 summarizes these devices. Details regarding instructions on the use of these devices can be viewed on the manufacturers’ websites.

FIGURE 30-7 Automated devices. A, EZ-IO. B, Bone Injection Gun.

(A, Courtesy Vidacare, San Antonio, TX; B, Courtesy Waismed, Kansas City, Mo.)

A review of the use of IO access in pediatric trauma patients up to 10 years of age reported successful placement of access in 28 out of 32 attempts in the prehospital setting and in the trauma care center (Guy et al., 1993). In this study, IO access was placed successfully by paramedics, nurses, and physicians with only one incident of minor extravasation of fluid and no long-term complications in the survivors. The commonest complication of IO access is subperiosteal infiltration that in most cases resolves spontaneously without further problems. The most feared complication—osteomyelitis—occurs in 0.6% cases (Rosetti et al., 1985). Other rare complications include fractures and emboli. Although few complications have been reported with this technique, it must be recognized that the high mortality in patients who require IO access prevents the assessment of long-term complications.

Disability (Neurologic Assessment)

A brief rapid neurologic evaluation is performed as part of the primary survey. It should include assessment of the patient’s level of consciousness and pupillary function. The AVPU method or the more detailed GCS should be performed. If AVPU is selected, the GCS calculation is performed during the secondary survey with a detailed neurologic examination. Periodic reassessment of the level of consciousness is necessary to detect neurologic deterioration caused by progression of traumatic brain injury (TBI), hypoxemia, or hypovolemia. Changes in mental status require prompt reevaluation of the ABCs. If they are adequately managed, then deterioration in mental status should be considered to be a result of TBI, prompting further brain imaging and consultation with a neurosurgeon.

Exposure

Exposure involves removing the trauma patient’s garments, usually with shears, to allow detailed physical examination and detect injuries. Rolling the patient, while maintaining cervical spine precautions, is necessary to identify injuries to the dorsal surface of the body that would otherwise be occult. Padding should be placed on the backboard at this time to prevent decubitus ulcer formation. This assessment should be rapid, and the patient should be covered in warmed blankets or with a warming device to prevent hypothermia. In addition, IV fluids should be warmed, and the room temperature should be raised. This is especially important in small children, who are more prone to hypothermia because of their larger surface area-to-volume ratio.

Secondary Survey

Fasting Duration

It is common practice to consider all trauma patients at risk for aspiration regardless of the time of last oral intake. The rationale for this approach is that major injury, the presence of pain and anxiety, and the administration of opioid analgesics delay gastric emptying. Additionally, bag-and-mask ventilation at the scene of the accident or in the emergency department that leads to gastric distention and the use of oral contrast solutions for diagnostic imaging studies may further increase the risk for aspiration. Indeed, previous investigators have demonstrated that patients who come for emergency surgery are at five times the risk for aspiration compared with those who undergo elective surgery (Olsson et al., 1986). Other investigators reported a 17% incidence of vomiting and 3% incidence of aspiration in 60 children younger than 19 years of age who required emergency ETI after they sustained a severe traumatic injury (Nakayama et al., 1992). Interestingly, residual gastric volume has been previously found to have a greater correlation with the interval from oral intake to injury than with actual fasting interval (Bricker et al., 1989).

These data raise two questions regarding the management of anesthetic induction for the trauma patient: whether it is possible to predict the safe interval between oral intake, injury, and induction of anesthesia, whether imposing a fasting duration once the trauma has already occurred offers any benefit in terms of reduction in aspiration risk. Goodwin and Robinson (2000) surveyed 167 practicing anesthesiologists in the United Kingdom regarding their practice in three different scenarios after a forearm fracture in a child. Approximately one third of the respondents did not believe there was any benefit in delaying the procedure and would perform a RSI and ETI regardless of the fasting duration, whereas almost two thirds of the respondents would delay the procedure if it was not emergent and then use a LMA or face mask as they would for elective cases. Such variability in clinical practice related to the management of the trauma patient is likely because of the difficulty in predicting a safe interval between oral intake, injury, and induction of anesthesia with regard to aspiration risk. A conservative and practical approach would be to proceed with surgery when an operating room becomes available and do a RSI and intubation if airway difficulty is not anticipated.

Induction of Anesthesia

Anesthetic induction techniques should be individualized according to the nature of the injuries, whether the airway has been secured before arrival in the operating suite, anticipated airway difficulty, hemodynamic status of the patient, and the presence of ongoing hemorrhage. The child with head trauma merits special consideration because of the risk of increased ICP during induction of anesthesia. Selection of an induction technique in these patients must be made with the goal of avoiding secondary brain injury. IV-induction agents such as thiopental, propofol, or etomidate may be preferred because of their beneficial effects on ICP and cerebral oxygen consumption (CMRo₂). On the other hand, a child with an anticipated difficult airway may be better managed with an inhaled route of induction so that spontaneous respiration is assured. Induction of anesthesia in a patient with dehydration or hypovolemia may lead to cardiovascular collapse. It is therefore imperative to have adequate IV access and rehydrate these patients before induction of anesthesia. A brief description of commonly used induction agents and pitfalls with the use of each in a child with trauma follows. Additional details about anesthetic agents are described in Chapters 7 and 22, Pharmacology of Pediatric Anesthesia and Anesthesia for Neurosurgery.

Maintenance of Anesthesia

Selection of agents for maintenance of anesthesia should be based on the nature and duration of the proposed procedure; the extent of injuries; the child’s ventilatory, hemodynamic, and neurologic status; and whether postoperative mechanical ventilation is anticipated. In hemodynamically stable patients, a standard general anesthetic may be used, including volatile agents, opioids for postoperative pain relief, and muscle relaxants as needed to provide good operating conditions. On the other hand, a technique using an opioid, a hypnotic, muscle relaxant, and oxygen may be more suitable in a child with unacceptably low blood pressure and who may not tolerate the negative inotropic effects of potent volatile anesthetics. In such cases, fentanyl or sufentanil are the preferred opioids, because they do not significantly alter hemodynamic parameters especially blood pressure. In children whose injuries are not completely determined, nitrous oxide is best avoided because it may diffuse into closed air spaces such as the pleural cavity.

In a child with severe head trauma, efforts must be directed at preventing secondary brain injury and protecting the injured brain from further ischemic injury by selecting anesthetic techniques that maintain blood pressure while reducing ICP. All volatile anesthetics cause cerebral vasodilation, which has been correlated with increasing minimum alveolar concentration (MAC) in children (Vavilala and Lam, 2002). Isoflurane affects CBF and cerebral autoregulation to a lesser extent than halothane. Sevoflurane offers greater advantages in that CBF velocities do not increase significantly with less than 1 MAC, and cerebral pressure autoregulation is maintained up to 1.5 MAC sevoflurane (Gupta et al., 1997; Monkhoff et al., 2001). For these reasons, sevoflurane may be the preferred volatile anesthetic for the child with TBI, and it would be prudent to limit its use to 1 MAC.

Opioids (fentanyl, sufentanil, or remifentanil) are often administered as intermittent bolus doses or infusions to supplement volatile anesthetics, for postoperative analgesia, and as additional measures to lower ICP. However, increased ICP has been reported in an 11-year-old child with closed head injury after bolus doses of fentanyl that responded to hyperventilation and barbiturates (Tobias, 1994). In addition, recent studies in adults have reported a transient but significant increase in ICP accompanied by a decrease in MAP and CPP after bolus doses of morphine, fentanyl, sufentanil, and alfentanil (Albanese et al., 1999; de Nadal et al., 2000). The exact mechanism of these changes remains unknown; however, impaired cerebrovascular autoregulation and direct cerebral vasodilatory effects of opioids have been implicated. Such effects may have significant implications in the management of the child with traumatic head injuries; however, until additional data become available, the judicious use of opioid infusions with careful monitoring of hemodynamic parameters is recommended.

Monitoring

In addition to routine monitors, placement of invasive monitors, including arterial and central venous catheters and a urinary catheter must be placed in the child with extensive injuries and in children with head trauma. Placement of monitoring catheters should be efficient, and in some cases when the surgery must be performed expeditiously, invasive catheters may need to be placed when the procedure is already underway. An arterial catheter is invaluable for continuous blood-pressure monitoring and frequent monitoring of arterial blood gases, acid base status, electrolytes, and serial hematocrits. Central venous pressure monitoring is useful when large fluid shifts are expected and in cases of rapid ongoing blood loss. Urine output is an important measure of fluid status. Additional monitors, such as ICP monitors in cases of head trauma and somatosensory evoked-potential monitors as a measure of spinal-cord function in a child with spinal injury, must be individualized. In some cases of chest trauma, transesophageal echocardiography may be useful in diagnosing injury and guiding surgical repair.

Temperature Maintenance

Temperature maintenance is an important part of trauma resuscitation but one that is often overlooked because of other priorities, such as resuscitation. In the traumatized child, several factors contribute to ongoing heat loss, including exposure to a cold environment at the scene of the accident, large open wounds, rapid infusion of fluids and blood, and exposure of body cavities during operative procedures. Hypothermia has significant deleterious effects that may hinder resuscitative efforts. Such effects include myocardial dysfunction, cardiac irritability, and dysrhythmias. Other effects of hypothermia include acid-base disturbances, a leftward shift of the oxyhemoglobin dissociation curve, and coagulopathies that can exacerbate ongoing hemorrhage. Continuous temperature monitoring in all pediatric trauma patients is therefore essential. Every effort should be made to maintain temperature, including the use of warm IV fluids, keeping the child covered once initial evaluation is completed, maintaining a warm environment, use of radiant warmers for infants, and forced-air warming devices.

Fluid Resuscitation

Shock is defined as a metabolic demand that exceeds either oxygen supply or oxygen delivery (Rasmussen and Grandes, 1994). When a child who has sustained multiple injuries arrives for surgical intervention, the fluid status must be assessed quickly (before induction of anesthesia) based on a physical examination as well as a history of fluid resuscitation received before arrival in the operating suite. The anesthesiologist must be prepared to continue the resuscitation in case of ongoing blood loss or third spacing of fluid. The goals of fluid resuscitation should be to maintain normovolemia as well as the osmolar and oncotic pressures in the intravascular space. Isotonic crystalloid solutions such as lactated Ringer’s solution or normal saline are most commonly used in the initial stages of resuscitation. Hypertonic saline solutions have also been used in this setting, based on the premise that they increase serum osmolality and thereby maintain intravascular volume for longer periods and with smaller volumes administered than isotonic solutions (Rasmussen and Grandes, 1994). However, the data that support these arguments are inconclusive and further work in this area is needed. The decision to administer glucose-containing solutions must be based on serial blood-glucose values (Sharma et al., 2009). This is of greatest importance in the presence of head trauma, because elevated blood-glucose levels have been found to correlate significantly with indicators of the severity of brain injury and poor neurologic outcomes in children with severe brain injuries (Michaud et al., 1991).

Colloid solutions such as 5% albumin and hydroxyethyl starch have also been used for fluid resuscitation. Hydroxyethyl starch may exacerbate existing coagulopathy by interfering with platelet function, decreasing fibrinogen activity, and interfering with factor VIII. It is therefore unsuitable for the pediatric trauma patient. The purported benefits of colloid solutions include their ability to increase colloid oncotic pressure and prolonged maintenance of intravascular volume and smaller volumes required compared with crystalloid solutions (Rasmussen and Grandes, 1994). For these reasons, colloids may also be beneficial in children with head trauma because the smaller volume of fluids administered may reduce the likelihood of cerebral edema. One of the major concerns with the use of colloids has been the cost. In most patients who require massive fluid resuscitation, the cost of using colloids to supplement crystalloids may be justified. The Saline vs. Albumin Fluid Evaluation (SAFE) study, a randomized, controlled trial conducted in 16 intensive care units in Australia and New Zealand, concluded that albumin and saline should be considered clinically equivalent treatments for volume resuscitation in intensive care patients (Finfer et al., 2004). Further discussion of crystalloid and colloid use in pediatrics has been reviewed by Bailey et al. (2010).

Blood Product Transfusion

The primary purposes for transfusion of blood products in a pediatric trauma patient are to maintain oxygen delivery and to ensure hemostasis (see Chapter 36, Systemic Disorders). Packed red blood cells are required when oxygen-carrying capacity is inadequate to meet tissue demands and metabolic rate. Losses of up to 40% of blood volume can usually be replaced with isotonic crystalloid solutions or colloids without physiologic signs of inadequate oxygen delivery (Solheim and Wesenberg, 2001). When estimated blood volume losses exceed 40%, the decision to transfuse blood should be based on an overall assessment of the patient, including the hemodynamic status, the extent of ongoing blood loss and the underlying comorbidity. Some children may require blood transfusion with blood-volume losses of less than 40% if the blood loss has been rapid or if they have significant underlying medical conditions such as congenital cyanotic heart disease or blood dyscrasias. Although there can be no fixed numeric transfusion trigger in all trauma patients, Box 30-4 presents formulas that may be used as general guidelines to calculate allowable blood losses (Rasmussen and Grandes, 1994). Blood banks in most centers supply blood components rather than whole blood. The primary advantage of component therapy is more efficient and cost-effective use of resources by eliminating the transfusion of unnecessary components and making components from a single blood donation available to several patients. It also permits improved preservation of individual components (Table 30-7).

Head Injuries

TBI is the leading cause of morbidity and mortality resulting from trauma in children (Langlois et al., 2006). A TBI is caused by a blow or jolt to the head or a penetrating head injury that disrupts the normal function of the brain. Severe TBI often includes an extended period of unconsciousness or amnesia after the injury and is associated with the poorest outcomes (Adelson et al., 2006). Among children ages 0 to 19 years, TBI results in an estimated 7441 deaths, 62,000 hospitalizations, and 564,000 emergency-department visits annually (CDC, 1999). The leading causes of TBI in children are falls (39%) and motor-vehicle crashes (11%) (Fig. 30-8) (National Center for Injury Control and Prevention, 2005).

FIGURE 30-8 A, CT scan of the head a 15-year-old girl who fell off a motor vehicle while “car surfing.” Note the skull fracture, epidural hematoma, and the midline shift of the brain before surgery. B, Same patient during craniotomy after drainage of her epidural hematoma; she was noted to have cerebral contusions in the area of her injury.

Most preventable deaths and deficits after pediatric head injury are secondary to subsequent complications, including diffuse brain swelling and resultant elevated ICP, which are independent predictors of mortality (Jurkovich et al., 2004). Poor CPP, often related to elevated ICP and trauma related systemic hypotension, are also known to worsen outcome. See related video online at www.expertconsult.com.

Problems that result from TBI, such as diminished cognition and memory loss, are often not visible (Langlois et al., 2006). TBI can cause a wide range of functional changes that affect thinking, sensation, language, and emotion (CDC, 2003). TBI can also cause epilepsy and increase the risk for conditions that become more prevalent with age, such as Alzheimer’s or Parkinson’s disease (CDC, 2003). The impact of a TBI not only affects the child but the whole family. The two age groups at highest risk for TBI are children between the ages of 0 and 4 years and those between 15 and 19 years old. Males are 1.5 times more likely to sustain a TBI, and African Americans have the highest death rate from TBI. The outcomes and the treatment of children with a severe TBI remain a great challenge. The CDC estimates that at least 5.3 million Americans, approximately 2% of the U.S. population, currently have a long-term or lifelong requirement for assistance with daily living activities as a result of a TBI (Langlois et al., 2006). Each year, 56,000 children are discharged home with a permanent disability from a TBI, whereas another 5000 require intensive inpatient rehabilitation facilities. The economic burden of head injury to patients is substantial—estimated to be $56.3 billion, with the highest rate of injuries reported among the lower socioeconomic classes. TBI is the most common reason a child requires intensive care or develops a significant life-long disability.

There are important differences between a TBI in a child compared with a TBI in an adult (Luerssen, 2006). Because of the smaller body mass of children, the energy received from an injury (e.g.,. fall) results in a greater force applied per unit body area and to a body with less fat and less connective tissue that might absorb or diminish the energy. The brain of a child is also anatomically different from that of an adult; the subarachnoid space is smaller and offers less protection because there is less buoyancy. Thus, head momentum is more likely to impart parenchymal structural damage. Additionally, the brain is proportionately larger to the rest of the body and therefore a bigger target for injury. Children are particularly susceptible to the effects of secondary brain injury that may be produced by hypovolemia with reduced cerebral profusion, hypoxia, seizures, or hyperthermia. Alternatively, the young child with open fontanels and mobile cranial suture lines is more tolerant of an expanding intracranial mass lesion. Overall, children with a TBI have better outcomes than adults with a TBI.

Monitoring of ICP and CBF are essential in the head injured patient. ICP can be measured with intraventricular catheters, subarachnoid bolts, and epidural sensors. CBF can be assessed by Doppler, and brain metabolic demands can be evaluated with the use of internal jugular bulb catheters and near-infrared measurements of mixed venous oxygen saturation Sv-o₂ (see Chapter 22, Anesthesia for Neurosurgery). In general, efforts are made to insure adequate venous drainage (30-degree, head-up position), adequate oxygenation, avoidance of hypotension, and maintenance of slight hypocarbia (partial pressure of CO₂ [Paco₂] of 35 to 38).

At present, there is insufficient evidence-based medicine to provide standards of care in children with head injuries. However, a number of interventions have become mainstay as guidelines for treatment of the pediatric patient with head injuries. Most of these guidelines emanate from data in adults (Krantz, 1996; Gupta et al., 1997). An algorithm for the treatment of intracranial hypertension in children is shown in Figure 30-9.

FIGURE 30-9 The University of Michigan C. S. Mott Children’s Hospital algorithm for the treatment of intracranial hypertension in children.

(Courtesy C. S. Mott Children’s Hospital, Ann Arbor, Mich.)

Primary and secondary brain injuries are the major variables affecting outcomes. Primary injury describes the immediate disruption of neuronal, axonal, and supportive structures and vascular tissue and is essentially untreatable except by prevention (Luerssen, 2006). If the primary injury is not fatal, it triggers a cascade of intracellular and extracellular biochemical changes that can augment and accelerate the injury. This is known as secondary injury. Secondary injury produces new damage to the tissue at the primary injury site as well as in other areas of the brain (Luerssen, 2006). For example, hypermetabolic responses related to neuronal tissue injury occur that may outstrip local or regional substrate supply. Ischemia is the final pathway that produces brain-tissue damage and poor clinical outcomes; it is caused by hypoxia, hypotension, seizures, hyperglycemia, and hyperthermia. Despite a lack of systematic evidence, there has been a trend toward improved outcomes by aggressively treating ischemia. The key features of such therapy are support of MAP, reduction of ICP to ensure CPP, and surgery for compressive lesions (such as epidural hematoma) (Alberico et al., 1987; Downard et al., 2000). Figure 30-10 depicts changes in MAP, CPP, and ICP with age. In addition to basic interventions of normal-to-slight hypocapnea, adequate perfusion pressure and oxygenation, the use of hyperosmolar therapy and hypothermia to treat increased ICP has been examined. Table 30-8 lists some of the commonly used interventions (excluding hyperosmolar therapy, discussed in the next section) and the strength of supporting data.

FIGURE 30-10 Age-related increases in MAP, CPP, and ICP.

Cervical Spine Injuries

Cervical spine injuries remain one of the most devastating consequences of trauma. The incidence of cervical spine injury in pediatric trauma patients is low (1% to 2%), but the morbidity and mortality of these injuries are substantial. Kokoska et al. (2001), in a 5-year review from the National Pediatric Trauma Registry, reported a 1.6% incidence (408 out of 24,740, 17% mortality) of blunt cervical spine injuries. Injury patterns were different by age with children less than 10 years of age having a higher incidence of C1-C4 injuries vs. C5-C7 injuries (85% vs. 57%; p < .01) (Kokoska et al., 2001). Brown et al. (2001) reviewed the experience of 103 pediatric cervical spine injury patients over 9 years and found an 18% mortality rate. Subluxation of cervical vertebrae and odontoid fractures are more common in children (Fig. 30-11). See related video online at www.expertconsult.com.Distinguishing between a true subluxation and a pseudosubluxation in a young child can be challenging (Table 30-9). In 40% of children younger than 7 years of age, pseudosubluxation-anterior displacement of C2 on C3 less than 3 mm is a normal variant. Pseudosubluxation becomes more prominent when a child is supine (i.e., on a backboard). Another normal variant is an increased distance between the dens and the anterior arch of C1 found in 20% of children (Bohn et al., 1990). The combination of a child’s proportionately large head, elastic interspinous ligaments, and horizontal position of the vertebrae is thought to explain both the pattern of injury and the normal variations (Bohn et al., 1990).

FIGURE 30-11 A, MRI of a 22-month-old boy who fell off of a grocery cart onto his head. Arrow points to a nondisplaced fracture of the odontoid process in C2. B, Posterior view of C2 odontoid (dens) fracture. C, Lateral cervical spine radiograph of a 9-year-old involved in an automobile accident. Arrow points to an occipitoatloid (C1) dislocation. The patient died.

(B, From Netter F, editor: The Ciba collection of medical illustrations, vol 1: nervous system, Summit, NJ, 1953.)

TABLE 30-9 Upper Normal Limits of Cervical Spine Measurements in Adults and Children

	Adults	Children
Predental space	2.5 mm	4 to 5 mm
C2-3 override (flexion)	3 mm	4 to 5 mm
Prevertebral space	7 mm	to AP distance (extension) vertebral body

From Eichelberger MR: Pediatric trauma prevention, acute care, rehabilitation, St. Louis, 1993, Mosby.

Current issues that relate to pediatric cervical spine issues include identifying who needs to be imaged, how that imaging should be performed, and what combination of imaging and clinical examination constitutes a “cleared” cervical spine.

One method to reduce excess and unnecessary imaging is to use Clinical Decision Rules (CDRs). CDRs are developed to reduce the uncertainty of medical decision-making by standardizing the collection and interpretation of clinical data (Laupacis et al., 1997; Stiell and Wells, 1999; McGinn et al., 2000).

The National Emergency X-Radiography Utilization Low Risk Criteria (NLC) and the Canadian C-spine Rule (CCR) are two clinical decision rules developed to assist with cervical spine imaging after a trauma. Their goals are twofold: to identify patients at risk for a cervical spine injury and to reduce unnecessary imaging. The NLC criteria were developed and then validated in a mixed population of adults and children; however, only 9% of the patients were considered pediatric.

Studies have shown that although CCR and NLC criteria may reduce the need for cervical spine imaging in children 10 years old and younger, the criteria are not sensitive or specific enough to be used as currently designed (Viccellio et al., 2001; Stiell et al., 2003; Ehrlich et al., 2009). Neither clinical decision rule is performed at a high enough level to be used with confidence.

A second issue pertaining to cervical spine clearance revolves around determining which imaging modality to use to assess the cervical spine. Physical examination in alert patients has been proven to be reliable; however, controversy remains as to which imaging studies to perform on pediatric CSI patients when the patient has altered mental status or is intubated (Marion et al., 2008). Most data for the diagnosis of cervical spine injures have been derived from the adult literature. Griffen et al. described a retrospective study of 3,018 blunt trauma patients of whom 1199 (40%) were at risk for CSI (Griffen et al., 2003). Plain radiographs (3 views) and a CT cervical spine were performed. Of 116 patients with spinal injuries, there were 41 patients whose injuries were missed by plain films, but diagnosed by CT scan. More significantly, all of the 41 patients with missed injuries required treatment. A prospective study from Schenarts et al. (2001) compared the use of cervical-spine CT for the upper cervical spine (C₁-C₃) to plain films for adult trauma patients with altered mental status. Plain films identified only 54% of the injuries to the upper cervical spine in their series, compared with 96% diagnosed with CT scan. In contrast to the previous study, a different study found that three injuries seen on plain films were missed on CT, including one atlantooccipital dislocation in a patient with quadriplegia and two subluxations (Schmoker et al., 1992).

Rana et al. (2009) recently examined their experience with cervical-spine imaging using either CT or conventional films in cervical-spine clearance with intubated children. Plain radiographs missed injuries and had a significantly lower sensitivity compared with CT scan (61% vs. 100%). In addition, all patients who remained intubated for 24 hours received flexion and extension views, but no new injuries were detected. This result is supported in adult trauma literature; Insko et al. (2002) reviewed the use of flexion-extension films and demonstrated that 30% of the patients were unable to flex and extend adequately, leading to further imaging with CT or MRI. They concluded that these films are of limited diagnostic utility in the acute setting for evaluation of the cervical spine at a Level I pediatric trauma center (Insko et al., 2002; Rana et al., 2009).

Radiation exposure from CT scans is an important concern and driving force for the rational use of imaging in children. Adelgais et al. (2004) examined this issue in a prospective study reviewing the differences of radiation exposure between conventional radiographs and CT. In this study, the patients receiving CT had 1.25 times the effective radiation dose compared with plain films. However, the comparison groups were markedly different; the patients undergoing CT scan were significantly more ill than the patients who only received plain films. The CT group had higher numbers of abnormal head CT (41% vs. 20%), intubation (20% vs. 2%), and transfer to intensive care or operating rooms (46% vs. 23%), compared with the group only receiving plain films. Keenan et al. (2001) examined this issue from another perspective by reviewing the number of excess radiographs required to clear the cervical spine of injury compared with the use of CT scan. Patients who underwent CT of the neck on initial evaluation had significantly fewer repeated radiographs than patients without cervical-spine CT (mean, 2.1 ± 2.6 vs. 3.6 ± 2.7 repeated radiographs, p = 0.04). The amount of radiation exposure was also determined, and the authors concluded that the children with initial CT did receive a higher effective dose (p < 0.001); however, patients with a GCS of less than 8 received equivalent doses compared with those not undergoing early cervical-spine CT (p=0.15) (Keenan et al., 2001). Although radiation exposure is increased in patients undergoing CT imaging, the actual increased risk of cancer above the natural background rate is low at 0.25% (Bier, 1990).

Child Abuse

The CDC defines child maltreatment as any act of commission (abuse) or omission (neglect) by a parent or other caregiver that results in harm, potential for harm, or threat of harm to a child. Unfortunately, such harm perpetrated on children permeates society. Even worse, child abuse remains to a great extent concealed or inadequately appreciated. Many experts believe that the number of annual referrals filed probably represent less than one third of actual cases; in fact, some sources describe as much as a tenfold difference between victim or parental self-reports and official estimates of incidence (Coté et al., 2009; Gilbert et al., 2009). Whether participating in the resuscitation and stabilization of a victim of acute pediatric trauma or assisting in the serial procedures required of more chronic medical conditions, anesthesiologists should step back and view their pediatric patients in a larger context that includes the possibility of child maltreatment.

Epidemiology

The U.S. Department of Health and Human Services collects and publishes data on child abuse in an annual report (U.S. Department of Health and Human Services, 2006). Of the estimated 3.2 million referrals made to Children’s Protective Services (CPS) in 2007, 61.7% progressed to investigation or assessment. Approximately 25% of these investigations indicated or substantiated child maltreatment, leading to an estimated 794,000 children being counted as victims of abuse or neglect. The most recognized forms of maltreatment are generally categorized as physical abuse, sexual abuse, psychological or emotional abuse, and neglect. During 2007, 59% of the victims suffered from neglect, whereas 11% were physically abused, 8% were sexually abused, and 4% were psychologically maltreated. Maltreatment is inversely related to age, as victimization is most prevalent among children under 1 year of age.

Perhaps even more disturbing is the high recidivism rate for perpetrators, because maltreatment often develops into a repetitive pattern of offense. In one study, 75% of victims younger than 24 months of age demonstrated evidence or history of previous trauma or injury (Kellogg, 2007). Children who are prior victims are 96% more likely to experience a repetition of abuse compared with those who are not previous victims. These trends are also inversely related to age, as adolescents from 16 to 21 years old are 51% less likely to experience recurrence compared with infants (U.S. Department of Health and Human Services, 2006).

Clearly, the most tragic outcome associated with maltreatment is a child’s death. Such cases often go unrecognized and unreported because of poor coordination and communication between various agencies, as well as a lack of standardized terminology and investigation techniques. Nevertheless, in 2007, 1760 children’s deaths were attributed to maltreatment. Again, the most fragile and defenseless children endure the greatest risk. Children younger than 1 year old accounted for 42% of fatalities, and 76% of the deaths attributed to maltreatment were associated with children under 4 years old (U.S. Department of Health and Human Services, 2006).

Many of these children have already been seen by various branches of the health care system, but the practice of pediatric anesthesiology provides a further level of screening for some of the most at-risk children. Children with disabilities, chronic illnesses, or special needs are at an increased risk for maltreatment and make up 8% of all child abuse victims (U.S. Department of Health and Human Services, 2006). It has been observed that compared with children who are not disabled, children with disabilities are 3.76 times more likely to be neglected, 3.79 times more likely to be physically abused, and 3.14 times more likely to be sexually abused (Hibbard and Desch, 2007).

Although child abuse affects children of all ages, genders, cultures, and socioeconomic strata, certain children are statistically predisposed to increased risk. Some of the indicative characteristics include the child’s age (infants and preschoolers) and whether the child was born premature, is developmentally delayed, possesses special health needs, or succumbs to behavioral problems. Also, given that greater than 80% of documented abuse cases are perpetrated by parents or caregivers, these victims’ home or care environments demonstrate other clear risk factors (Box 30-5) (Hornor, 2005; Kellogg, 2007). Obviously, bias should be avoided, because these risk factors in isolation do not indicate child abuse. Such characteristics should instead aid and guide further investigation as well as prevention strategies and treatment plans.

Box 30-5 Risk Factors for Child Abuse

Parental or caregiver characteristics

Poverty

Unemployment

Lack of or minimal education

Social isolation

Single parenting or unrelated caregivers

Caregiver history

Substance abuse

Domestic violence

Mental illness

Their own neglect or abuse as children

For the anesthesia provider, a patient’s preoperative or preprocedure evaluation is often streamlined and focused on the impending operative physiology and risks; however, certain information should raise a level of concern for possible abuse:

• There is a lack of or inadequate explanation for the injury or condition.

• The details of the explanation change.

• There are differing accounts from witnesses.

• The explanation does not fit the pattern, age, or severity of the injury.

• The explanation does not fit the child’s developmental capacity.

• There are delays in seeking medical attention for the injury or condition.

• The concern expressed by caregivers is inappropriate for the condition or injury sustained.

It must be noted that children may be naturally prone to injury as they test the limits of their autonomy. Most childhood injuries are not the result of maltreatment; hence, common sense should apply when considering each child’s ability to sustain observed bruising or trauma (Fig. 30-12). The same injury in a mobile 4-year-old child may be diagnostic of physical abuse in an infant. Likewise, outside of infancy, superficial injuries or bruising should be consistent with the hazards of daily living activities. Bruising to skin over bony prominences (e.g., shins, knees, forehead, chin, and forearms) is more consistent with normal activities, whereas injuries to protected, padded areas (e.g., neck, abdomen, buttocks, genitalia, cheeks, and ears) should always be suspect.

FIGURE 30-12 A, Severe bruising with underlying muscle damage. This toddler was covered from head to toe with severe looped-cord contusions and lacerations. He had secondary myoglobinuria, necessitating intensive care management to prevent renal failure. B, This boy was struck with a chain, leaving a clear imprint of the links. C, A deep, circumferential rope burn on the wrist with considerable edema and early skin breakdown of the hand is seen in this infant, who was tied to the siderails of her crib. D, This circumferential cord burn was the result of an attempted strangulation. E, Multiple retinal hemorrhages are seen on funduscopic examination of this infant who was a victim of the shaken-baby syndrome. Subdural hematoma and multiple metaphyseal “shake” fractures are typical associated findings. F, A superficial laceration is seen over the left eyelid. G, A deeper stab wound over the right flank penetrated into the subcutaneous tissue.

(From Zitelli BJ, Davis HW, editors: Atlas of pediatric physical diagnosis, ed 4, St. Louis, 2002, Elsevier.)

Anesthesiologists are perfectly positioned to routinely complete a full-body survey in the course of each patient’s positioning in the operating room and for anesthesia. The key is to recognize a pattern or constellation of physical findings that may point to possible abuse or neglect. Some compelling or indicative findings are noted in Table 30-10. Complete documentation of injuries, including diagrams and photographs, is paramount to facilitate peer review as well as CPS and law enforcement intervention and court proceedings.

TABLE 30-10 Physical Examination Findings Associated with Maltreatment

General	Notes
Extensive dental caries	Examples of general neglect
Severe diaper dermatitis
Derelict wound care
General uncleanliness
Failure to thrive (may manifest in nutritional neglect)
Anxiety and reluctance when the child is asked about abuse
Skin	In groups or patterns, especially around soft, protected, padded areas
Bruises
Bites
Lacerations
Burns • Immersion: genital, “stocking,” or “glove” pattern from dipping the afflicted body part • Contact: from cigarettes, irons, curling irons, etc. (especially branding pattern) • Chemical: ingestion (may indicate neglect or abuse)

  Head Bruising or injury to ears, nose, or mouth   Fontanel: fullness or asymmetry (may indicate step-off fracture or intracranial pathology)   Eyes not tracking: may indicate head injury   Retinal hemorrhage (by funduscopic exam)   Thorax/Abdomen Any significant bruising or trauma, especially in infant or nonambulatory child   Skeletal Specific fractures carry higher specificity for abuse (especially in infants)

• Posterior ribs fractures

• Sternum fractures

• Scapular fractures

• Spiral long-bone fractures (from torsion or twisting of extremities)

• Metaphyseal fractures (from whole traction forces on limbs; may be seen with shaking)

If abuse is suspected, other tests may be appropriate to assess for additional injuries or to elucidate other medical etiologies for injuries. These diagnostic studies may be more easily determined in consultation with child-abuse specialists. Regardless, the younger the child or the more severe the injury, the greater the need exists to evaluate for other injuries. And if one child is a suspected victim, other siblings or children in contact with the caregivers should also be assessed.

Management Issues

Laws mandating the report of suspected abuse and neglect exist in all 50 states. Proof is not required before reporting, and good-faith reporting has immunity in some form or another (Kellogg, 2007).

Almost all states have what are nominally called child-protection teams. Such teams are composed of specialists who are well versed and trained in the medical, social, and legal aspects involved with these cases. The members of these teams are composed of social workers and physicians or medical providers with specific experience in the field. In fact, the American Academy of Pediatrics has recently implemented and accredited a new board subspecialty in Child-Abuse Pediatrics. Such child-protection teams should be available to concerned clinicians and typically work in conjunction with emergency departments, primary care practitioners, law enforcement agencies, and county CPS. These teams may offer onsite assistance and provide consultation on specific case questions.

In most tertiary care or academic centers, assistance may be readily available. However, in other health care facilities, there may not be any familiarity with these processes, so providers should solicit assistance from hospital social workers, administrators, or regional tertiary care centers. In the absence of any other source or after normal business hours, law enforcement and CPS are available to provide information and assure child safety. The U.S. Department of Justice also furnishes applicable information (http://www.ojp.usdoj.gov/ovc/help/ca.htm) and advertises a child-abuse hotline (1-800-4-A-Child) for referral information.

In order to recognize child maltreatment, anesthesia providers must understand and acknowledge that abuse and neglect occur on a regular basis, that such victimization occurs in any segment of society, and that maltreatment must be considered as part of the differential diagnosis when assessing any instance of childhood trauma.

Dog Bites

Dog-bite victims represent another common instance of facial injury in children. In a study by McHeik et al. (2000), the majority of children suffering dog bites (68%) were younger than 5 years of age. It was believed that the increased rate of dog bites in this age group was as a result of young children’s lack of caution and comprehension regarding animals, coupled with the fact that their heads and necks are on the same level as that of a dog. A study by Bernardo et al. (2002) revealed that younger children (younger than 6 years old) commonly sustained dog bites from their family dogs in their own homes and confirmed that these injuries typically involve facial structures. The goals of immediate surgical repair are to diminish scarring and decrease the rate of wound infection.

As with other trauma, facial injuries in children should be evaluated in the field or the emergency room during the primary survey. If there is any evidence of airway obstruction or respiratory insufficiency, the airway should be secured immediately with LMA, ETT via direct laryngoscopy, LMA, or surgery. As mentioned previously, special attention to cervical spine stability during direct laryngoscopy should be accomplished with inline stabilization of the neck. Once the airway is secured, a secondary survey can be performed to ascertain the extent of the facial injury and the need for surgical intervention.