#92404: Pediatric Abusive Head Trauma

Serious injuries that result in the death or debilitation of infants or young children are seldom the result of accidents. It is estimated that 95% of severe intracranial injuries and 64% of all head injuries in children 1 year of age or younger are caused by violence inflicted by parents or caretakers [10]. By contrast, the incidence of infant death by accidental fall or drop from low height (less than 4 feet) is less than 0.5 in 1 million [20].

Though widely known today as shaken baby syndrome (SBS), it is important for healthcare professionals to embrace the all-inclusive medical terminology pediatric abusive head trauma (AHT), not to detract from shaking as a mechanism of AHT, but to acknowledge the various mechanisms of deliberate injury to infants and children [1]. AHT is a spectrum, ranging from mild injury due to sub-lethal abuse, which can cause lethargy, irritability, poor feeding, and/or vomiting occurring for days or weeks, to the most severe injury leading to coma and/or death [1]. In fact, pediatric AHT is the leading cause of death and debilitation in children among all forms of physical abuse [2]. The symptoms and conditions associated with AHT have been found in children up to 7 years of age but occur most commonly between the ages of 2 and 9 months, coinciding closely with the peak of infant crying [38,40,70]. The great tragedy is that AHT is thought to be highly preventable with well implemented parental education programs and access to support networks and services.

Investigation of pediatric AHT began in the 1940s by Philadelphia physician John Caffey. Dr. Caffey published a seminal article in 1972 on whiplash-shaking of infants and what he termed "whiplash shaken infant syndrome." A variety of reasons for whiplash-shaking were found, but parents or caretakers would typically only admit to shaking infants or children to correct misbehavior [4]. At the time, "giving a child a good shaking" was an acceptable disciplinary technique. Through accumulated circumstantial evidence, the correlation was subsequently made between abusive treatment or violent shaking and injuries to the brain and its blood vessels, the vessels of the eyes, damage to the cervical spine, and long-bone injury, as it was often found that toddlers were shaken by their arms or hung upside down and shaken from their legs. Unfortunately, the widespread use of shaking to stop crying and as a form of discipline still persist to this day [51].

It is difficult to gauge the exact incidence of AHT for a number of reasons, including the criminal nature of child abuse. Historically, discrepancies in defining and coding pediatric head trauma and abuse (and subsequent inconsistency in national data sets) have led to poor estimates [51]. Additionally, a vast number of cases evade clinical detection. Based on hospitalization data from the 2000, 2003, 2006, and 2009 Kids' Inpatient Database (Healthcare Cost and Utilization Project), the incidence of detected AHT in the United States is approximately 40 cases per 100,000 live births for children younger than 1 year of age and 7 cases per 100,000 for children 1 year of age or older; this equates to approximately 1,200 to 1,400 detected cases per year [48,64,70]. This rate varies regionally due to socioeconomic factors as well as whether or not AHT prevention education programs are provided to parents and/or other members of society involved in the care and safety of children [11,48]. A 2022 analysis found that between 2006 and 2018 in the United States, there were 12,287 emergency department visits for AHT in patients younger than 5 years of age; the number of visits decreased 6.7% each year [74].

Specialists in pediatric abuse, neurology, and psychiatry consider undetected cases of AHT to be a major cause of behavioral problems, developmental disorders, and intellectual disabilities of unknown etiology (19% to 90%), although to date, this association remains unconfirmed [51,52,53,54,55,71]. The likelihood that most subclinical cases of abuse are never detected is high, as is the suspected number of AHT injuries that cause lesser degrees of behavioral/developmental symptoms, but the proof linking these symptoms to undetected abusive injuries may never be found.

Approximately 18% to 25% of severe AHT victims will ultimately die from their injuries, and 65% to 80% of survivors will become significantly disabled, resulting in great societal and financial costs [49,61,62,63,65,71]. The total economic impact of AHT is unknown, but overall child abuse-related costs are estimated to be $124 billion per year (in 2010 dollars) in the United States, with $18.6 billion attributed to AHT [51,65]. This figure is based on the conservative estimates of $210,000 in lifetime medical costs for each case of nonfatal abuse and $1.2 million for each abuse-related death. Lifetime costs for each AHT case are likely much greater ($300,000 to more than $7 million) due to the wide range of serious injuries and permanent disabilities (e.g., depression, learning disabilities, visual disabilities, seizure disorders, cerebral palsy, paralysis) [51,65]. Additionally, the lifetime costs related to undetected AHT are potentially massive. The direct medical costs of AHT admissions in the United States were estimated to be $70 million annually in 2015 [67].

AHT is not a phenomenon unique to the United States. One international study of child abuse found that the prevalence of shaking children 2 years of age or younger as discipline or punishment (equivalent to abuse) was between 20% to 63% in 9 of 16 communities sampled in Brazil, Chile, Egypt, India, and the Philippines [58]. Surveys of mothers in Canada and Holland revealed that 2% and 5%, respectively, agreed with a question that asked if "shaking is a good way to silence a crying infant" [42,51,59]. In Holland, 5.6% admitted to slapping, smothering, or shaking an infant because of his or her crying. A survey in North and South Carolina found that 2.6% of mothers routinely use shaking of infants and toddlers as discipline, a rate higher than was admitted by fathers [60].

Abusive behaviors associated with AHT include intentionally throwing, dropping, striking, and vigorously or violently burping, grabbing, pulling, pushing, shaking, or slamming an infant or young child. The majority of abuse-inflicted deaths and debilitations are the result of brain injury and cerebrovascular trauma combined with cervical spine injury and secondary injuries (e.g., hypoxia, ischemia, metabolic cascades) [1]. However, many instances of abuse-inflicted head injuries (e.g., lacerations, skull fractures) have been discovered without detectable neurologic complications [34].

As discussed, shaking is often given as the sole explanation of patients' injuries, most commonly attributed to misbehavior or persistent crying; one study found shaking to be the only admitted behavior in 68% of confirmed AHT cases [32]. This supports the belief that shaking is the primary mechanism of AHT and the widespread use of the "shaken baby syndrome" terminology. Shaking is commonly identified by abusers as being harmless; however, the American Academy of Pediatrics (AAP) notes that anyone witnessing an act of abusive shaking would recognize it as dangerous and likely to harm the child [1].

Some researchers believe that shaking alone cannot cause the pattern of injuries associated with "shaken baby syndrome" in an otherwise healthy infant, no matter how violent. These researchers assert that impact forces (from either soft or rigid surfaces), which are greater than shaking forces by a factor of 50, probably cause the majority of injuries in the AHT constellation [37]. However, this assertion is based on research using an adult primate model as an analogue for an infant brain, which is now thought to be an inaccurate comparison [76]. The abundance of research and literature supporting the causal relationship between violent shaking and pathologic findings suggests that shaking alone can cause sub-lethal injuries (e.g., concussion, minor intercerebral bleeding) and more severe injuries. Others have proposed that AHT is almost always a combination of shaking and impact forces.

Typically, there are no witnesses to the abuse in almost all AHT cases, so the exact mechanisms of various head traumas are without validation [1,57]. Violence is rarely volunteered as an explanation for a child's injuries, especially when there are no witnesses. When violence is suspected, many perpetrators (in the clinical setting; or through their legal counsel, in court) attempt to defend their actions by diminishing the severity of the incident. This can present problems for diagnosis in the emergency department and when prosecuting cases in court, as expert defense witnesses cite the lack of scientific evidence (i.e., lack of a model that accurately represents a human infant) that supports shaking alone as a cause of significant injury [57]. However, in one report of 29 judicial confessions in AHT cases, the perpetrators all described extremely violent shaking and 55% admitted repeated shaking from 2 to 30 times (average: 10 times), mostly because it stopped an infant's crying [56]. Daily shakings for weeks or months occurred in 20% of cases because of their admitted effectiveness in stopping crying episodes. Impact was confessed in 24% of cases. This report included data showing a similar pattern of injuries in 83 other judicial AHT cases where there was no confession of either shaking or impact [56].

In some presentations, there may not be marks or bruises to alert healthcare professionals to obvious physical abuse or to support the suspicion of abuse; this is often due to the abundance of soft household surfaces, such as pillows, sofas, or mattresses, that the child may be struck with or thrown onto. Although soft surfaces reduce obvious signs of abuse, they still cause sudden deceleration (e.g., a child being thrown into a crib or onto other furniture) or acceleration (e.g., hitting the head with a pillow) injuries [18]. The lack of marks or bruises may also be due to the possibility that the only mechanism of injury was extremely violent shaking while the child was wrapped in something that provided some cushioning from a forceful grip, such as a thick blanket.

Direct injury to the brain can be caused by skull deformation (e.g., from a blunt impact; rigid object) or rapid acceleration and deceleration forces upon the brain, or by the rotational forces that can result from rapid acceleration/deceleration. Rotational forces can cause a range of direct brain injuries, from concussion to diffuse axonal injury. Rotational forces can also damage the intracranial vascular system, greatly impacting mortality and morbidity related to hemorrhage and associated brain damage. However, early diagnosis and treatment of bleeding offers a better outcome than direct brain injury, which may have more ambiguous symptoms. Although many assume that a blunt force impact with cranial deformation would cause greater damage than shaking, insults from shaking or repeated slamming against or hitting with soft objects may be many times worse [4]. This is because blunt impacts cause focal damage from linear forces, whereas rotational forces cause diffuse, lethal AHT injuries [76,77]. The brain of an infant is softer than an adult's, with the consistency of unset gelatin; as such, they are more easily affected by rotation [18].

Whiplash-type neck injuries are also common in abused pediatric patients because their heads are proportionally larger and heavier than adults' and their neck muscles are not well developed. Therefore, the infant head's relative mass amplifies trauma, and no shock absorbing effect is available. Damage to the cervical spine is not uncommon in AHT cases.

Because the goal of this type of abuse is usually to stop the child from crying or otherwise aggravating the caretaker, the end result is that the sustained injuries are typically of a magnitude that causes the infant or toddler to lose consciousness for an extended period of time. This is often the presenting condition.

Aside from being comatose, a pediatric patient with traumatic head injuries of abusive origin may be convulsing, have altered consciousness, have impaired ability to suck or swallow, and be unable to track with eye movements, smile, or vocalize [1]. Skin or lips with a blue hue and lethargic eyes are signs of AHT. However, in many cases there may not be any physical clues to assist the assessment.

The most apparent clinical sign of AHT is a scalp laceration. Blunt force applied to the head will cause tissue disruption, leading to hemorrhage. In small children, uncontrolled hemorrhage may cause exsanguination attributable to the high density of the vascular bed in the scalp. In most patients, pressure dressings applied to the injury may slow or stop this loss of blood. In acute situations, hemostats and ligation or cautery of scalp vessels may be required to staunch the ongoing blood loss.

Some patients will present with bruises on the arms, legs, body, neck, and/or head; bleeding; and/or skull deformation. However, in cases in which there are no obvious physical signs of abuse, radiologic imaging may be helpful in diagnosing intracerebral hemorrhage or spinal injury [18,19]. Magnetic resonance imaging (MRI) and computed tomography (CT) will have limited usefulness in diagnosing certain conditions.

Injury to the neural tissue can vary in severity from a concussion or traumatic brain injury (TBI), the mildest form, to a contusion or a diffuse axonal injury, producing severe, profound coma. A concussion produces a temporary disruption of cerebral function, generally lasting less than one day.

Contusion injuries of the brain cause a more prolonged compromise of cerebral function than a concussion. On examination, the brain tissue is edematous and may have signs of localized hemorrhage. Loss of consciousness may occur for prolonged periods of time, and recovery outcomes may be unpredictable at the time of initial evaluation. Often, contusions occur as coup-contrecoup injuries. This injury emerges during rapid acceleration and deceleration of the head as occurs in a motor vehicle accident or in AHT. The coup injury occurs when the brain impacts the interior of the skull, leading to injury of the brain tissue at the site of impact. The contrecoup lesion can be of greater consequence; this lesion occurs on the brain at the side opposite from the initial impact. In other words, the brain has bounced inside the skull, causing damage to a number of areas in the brain. The greater the damage to the neural tissue, the greater the risk of severe brain injury.

When the head is injured, it is important to remember that two types of injury are possible: primary and secondary. Primary injury to the head and brain occurs at the time of trauma. Healthcare professionals have little control regarding the development of this type of injury, with the exception of education regarding prevention. Secondary injuries occur as a result of the trauma; examples include cerebral edema, bony fragments, and delayed vascular injury. Trauma care of injuries to the head and brain is directed at preventing or controlling the development of secondary injury.

Abusive head trauma can be divided into three categories of injury that may or may not appear concurrently: intracranial injuries, skull fractures, and spinal injury.

Immediate measures in trauma resuscitation for the patient include management of the airway, breathing, and circulation (ABC) while simultaneously determining if he or she has sustained any life-threatening injuries. In the child, these steps take on even more importance, as the loss of the airway can be rapid and result in devastating consequences. The limited size of the pediatric airway increases the risk of rapid deterioration and subsequent difficulties in management. All pediatric head trauma patients should be managed as if a spinal cord injury exists. This requires utilizing the jaw-thrust maneuver to open the airway while maintaining alignment of the cervical spine. Once positioned, the airway should be examined for debris, such as loose teeth, blood, or saliva that can be mechanically removed. As the ABCs are stabilized, the primary survey includes assessing the level of disability and requires a complete examination of the patient for rapid identification of any underlying injuries. The primary survey should take no longer than 5 to 10 minutes for assessment and stabilization.

It is important to remember that severe head trauma in infants and young children is rarely accidental. When no other explanation of severe trauma is clear (e.g., a vehicle collision) or the provided history does not match the child's developmental age, an investigation of abuse should be pursued [26]. The proper authorities should be notified, and specialists in ophthalmology, pediatric radiology, pediatric neurology, and/or pediatric neurosurgery and a child abuse pediatrician should be part of the treatment team [1]. It is also essential for all pediatricians and other healthcare professionals involved in the care of children to be aware of the more subtle signs of physical abuse and past head trauma and for pediatricians to be confident in properly examining children for abuse. Other healthcare workers who suspect child maltreatment should enlist the help of a pediatrician, preferably one that specializes in child abuse.

As noted, abusive injuries can range from minor bruises and lacerations to the more severe head trauma discussed in this course. Physical abuse is one of the most easily identifiable forms of abuse and is the type most commonly identified by healthcare professionals. Physical injuries that may be indicative of abuse include bruises/welts, burns, long-bone and rib fractures, abdominal injuries, lacerations/abrasions, and manifestations of central nervous system trauma. Bruises and welts are of concern, particularly those that appear on:

Patterns, such as shapes of the article used to inflict the bruise or welt (e.g., a cord, belt buckle, teeth, or hand), should be noted. Cigar or cigarette burns are common indicators of abuse, and they will often appear on the child's soles, palms, back, or buttocks. Patterned burns that resemble shapes of appliances, such as irons, burners, or grills, are of particular concern.

When examining patients, note bruises on the abdominal wall; any intestinal perforation; ruptured liver or spleen; and blood vessel, kidney, bladder, or pancreatic injury, especially if explanations for the cause do not make sense. Look for signs of abrasions on the child's wrists, ankles, neck, or torso. Lacerations might also appear on the child's lips, ears, eyes, mouth, or genitalia. Healing lacerations or scars on a child's head, especially when concurrent with other recent injuries, are cause for suspicion of abuse. All injuries should be documented and photographed.

According to the American College of Radiology, imaging the head in children with suspected abuse depends on the child's age and type of presentation. In children with skull fractures or clinical signs and symptoms of intracranial injury, an immediate noncontrast computed tomography scan of the head should be performed.

(https://acsearch.acr.org/docs/69443/Narrative Last Accessed: May 23, 2023)

Level of Evidence: 9 (Usually appropriate)

Fractures that result from abuse might be found on the child's skull, spine, ribs, hands, feet, nose, or any facial structure. These may be multiple or spiral fractures at various stages of healing. As discussed, mechanisms of injury that cause AHT, especially injuries from shaking, can cause multiple long-bone end fractures. It is recommended that certain high-risk groups with normal neurologic statuses, especially patients younger than 1 year of age, be selected for radiologic imaging of the head and a complete skeletal survey when no reasonable explanation of injuries exists. This includes those with rib fractures, multiple fractures, or facial injury [34]. This is because skeletal survey alone can miss more than 25% of cases of abusive head injury [34]. One study found that up to 30% of AHT is initially undiagnosed, and mean time to diagnosis is seven days, having potentially deadly consequences [7].

As discussed, retinal hemorrhages are characteristic of AHT cases, with more extensive bleeding associated with a greater extent of brain injury [1,39]. Retinal hemorrhaging has long been used to make a more definitive diagnosis of abuse.

The decision of whether or not to report suspected abuse is ethically challenging. Although healthcare professionals are ethically and legally obligated to report suspected child abuse, suspicion of abuse is somewhat of a judgment call, and certain biases may influence the decision to report. A 2008 prospective observational study found that clinicians did not report 27% of injuries considered "likely or very likely caused by child abuse" and 76% of injuries considered "possibly caused by child abuse" [23]. However, patients who had an injury that was not a laceration, who had more than one family risk factor, who had a serious injury, who had a child risk factor other than an inconsistent injury, who were black, or who were unfamiliar to the clinician were more likely to be reported. The AAP makes several useful recommendations regarding the diagnosis and reporting of AHT [1]:

The foundation of AHT prevention is believed to be an infant shaking awareness program implemented on a large scale, targeted to new parents and child caretakers. Information regarding infant crying and how to deal with potential stress and rage associated with crying and infant care are provided in this training. However, the effectiveness of AHT reduction programs is uncertain, with two large studies published in 2015 and 2017 questioning the positive outcomes of earlier trials [41,42].

Education regarding AHT provided to the mother and father (or father figure) before an infant is discharged from the hospital was shown to decrease the number of abusive infant-shaking cases by roughly half overall in one study [38]. The study initiated in 1998 in an eight-county region of western New York and involving all hospitals that provided maternity care began with the assumption that because parents are so often the perpetrators of abuse (roughly three out of four cases), an educational program targeted at new parents, rather than an awareness campaign to the general public, could more effectively reduce the incidence of AHT. The methodology incorporated assessment of parents' prior knowledge of the dangers of infant shaking (which proved to be high at 92%), education about normal infant behavior, alternative responses to persistent crying, obtaining a signed commitment statement from mothers (and whenever possible, fathers or boyfriends), and a follow-up seven months later with a random sampling of 10% of parents.

The nurse-parent interactions were kept brief; a five-minute teaching interaction was average. The rationale was that a less time-consuming interaction was more likely to be implemented by hospital staff and was acute enough to keep parents' attention while they were naturally preoccupied. Several low-cost (compared with the high cost of care for a child with head injuries) measures were implemented: brochures were handed out, posters were hung, and an 11-minute video was to be viewed before signing the commitment statement. At the follow-up, 94% remembered receiving information about shaking, primarily the written materials (98%), the commitment statement (92%), and the interaction with staff (89%). It became clear that most parents were not shown the video. The rate of AHT in the eight-county NY region during the six-year period before the study (41.5 cases per 100,000 live births) was used as a control; additionally, the rate of AHT in the state of Pennsylvania was used to bracket the control. With other factors reasonably accounted for, the incidence of AHT dropped almost in half, to 21 cases per 100,000 live births in the six-year study period, while rates in Pennsylvania remained essentially the same.

There are multiple studies documenting the success of infant shaking prevention programs, including the "Love Me…Never Shake Me" program in Ohio and "The Period of PURPLE Crying" program from the National Center on Shaken Baby Syndrome [41,42]. Analysis of these programs shows that information given at or shortly after birth, along with continued reminders at each office visit, is the most effective in reducing traumatic abuse. Considering that most parents already know that violent infant shaking is harmful, it is perhaps important for parents to make the association between this knowledge and their own infant and for parents to learn the particulars about self-care along with information about normal infant behavior. Key points of the AHT education program should include [42,43,44]:

In The Period of PURPLE Crying program, the PURPLE acronym that can be a useful reminder for new parents in the two- to five-month period after birth, when infant shaking is most likely to occur [44]. The acronym stands for [44]:

In light of the considerable decrease of AHT cases in New York due to the western NY pilot program, the same pilot program was initiated in Pennsylvania in 2002. State legislation was also passed in 2002 to require all birthing and children's hospitals to implement the Pennsylvania Shaken Baby Syndrome Prevention and Awareness Program by 2006 [45]. The original western New York program began implementing education in many of the region's pediatric offices in 2005, leading to an additional 10% decrease in AHT cases [45].

Despite these apparent successes, several failures have also been noted. A study published in 2017 found that a parent-targeted AHT prevention program failed to decrease the AHT hospitalization rate over a 10-year period [68]. The study included parents of almost 1.6 million infants in the state of Pennsylvania and compared the hospitalization rate to that of five other states without statewide AHT prevention campaigns. Although parents did demonstrate improved awareness of AHT, the incidence of AHT hospitalizations in Pennsylvania actually increased slightly (from 24.1 to 26.6, per 100,000 children), while those of the other states remained stable [68]. Only 20% of the parents saw both the brochure and video, and only 6% completed the entire intervention.

A 2015 North Carolina study (including more than 400,000 newborns) of The Period of PURPLE Crying intervention also failed to show a statistically significant reduction in AHT rates [69]. The study controlled for family's economic differences (e.g., unemployment, mortgage foreclosure) and used other state's AHT rates in the analysis. The authors discussed the potential impact of the deep economic recession that occurred during the study period, and they acknowledged that the particular economic risk factors for AHT might have evaded their analysis or that state-level economic factors could have been significantly different between North Carolina and states used for comparison [69].

All healthcare professionals involved in pediatric care have an ethical duty to advocate for AHT prevention programs at their place of practice. Only through preventative educational measures and early recognition of the signs and symptoms of abuse can the incidence of AHT be significantly reduced. It is necessary to educate expectant and new parents about stress relief techniques, normal infant crying behavior, and the importance of only leaving their children in the care of individuals whom they trust implicitly and ideally who also have completed AHT prevention education. When abuse is suspected to have occurred while in the care of another individual, parents should be encouraged to report the individual to the proper authorities. This is especially important when the abuse is thought to have occurred at a child-care center or during unlicensed home-based care. It is also vital to remember that the significant amount of attention given to SBS should not detract from awareness of abuse in older pediatric patients, who are also at an increased risk for AHT due to the propensity of violence inflicted to the head and face.