Written by: Dana Loke, MD (NUEM PGY-4) Edited by: Ashley Amick, MD (NUEM ‘18) Expert commentary by: Lauren Riney, DO
Non-accidental trauma (NAT) is a leading cause of pediatric traumatic injury and death. In 2014 alone, there were 1546 reported deaths from NAT and 3.6 million child abuse referrals submitted to Child Protective Services (CPS).  NAT is most commonly encountered in young children, but can occur at any age. The classic signs and symptoms of NAT will be reviewed here, but it is important to realize that occult injury is common. Compared with accidental pediatric trauma, patients with NAT have been shown to have higher injury severity scores, rates of intensive care unit admission, and mortality. Furthermore, the diagnosis of NAT is delayed in 20% of cases, increasing the risk of poor outcomes. Therefore, the Emergency Physician (EP) must maintain a high index of suspicion for NAT to prevent the grave consequences of missed diagnosis for the patient and any other children in the home.
Red Flags and Risk Factors
NAT is a frequently missed diagnosis, but there are some red flags and risk factors that should make the EP take pause and consider this diagnosis. Children at greatest risk are generally toddler and younger, and often come from dysfunctional family units. A recent study found that 97% of NAT cases have antecedent familial dysfunction, such as substance abuse (alcohol or drugs), psychiatric disorder, history of violence or incarceration, or child withdrawal.  Additionally, over 70% of reported NAT deaths in 2014 were in children under 3 years old. 
Injuries inconsistent with the caregiver’s history
Reported mechanism of injury is unexpected for the child’s developmental status (for instance, a 2 week old infant rolling off of a bed)
Age under 5 account for 81.5% of cases; children under 1 are most vulnerable 
Multiple medical conditions
Female parent (although males are more likely to inflict fatal NAT)
Poor social support
Unplanned or unwanted pregnancy
Poor prenatal care
Shorter birth intervals between children
Increased number of separations from the child in the first year
Depression and suicide attempts
Personal history of being abused as a child
Exposure to foster care or abandonment as a child
Engagement in criminal activity or corporal punishment as a child
Many other suspected risk factors have been studied. There is no consensus regarding whether a particular race is at greatest risk for NAT however black children have a greater risk of mortality from NAT.  Similarly, there is no consensus regarding socioeconomic status as it relates to NAT risk, but studies have shown that incidence of non-accidental head trauma and its severity rise during times of economic recession. 
Bruising is the most common manifestation of NAT but has low specificity. In any child presenting with bruising, it is imperative to note the location, shape and pattern of the lesion and ensure this is clearly documented. Bruising located over soft tissue areas such as the cheeks, neck, genitals, buttocks, torso, and back, are more likely to represent NAT than bruises over bony prominences.  The shape of the bruise should be considered as well, since the bruise often reflects the shape of the causative object. Common objects used to inflict injury include belts, cords, shoes, kitchen utensils, hangers, and teeth.  Additionally, patterned bruises should raise suspicion for NAT since they generally do not occur with accidental trauma. Lastly, any bruising in non-mobile infants is suspicious for NAT as well. 
Burns occur in 8-12% of NAT cases.  The most common types of burns from NAT are scald burns and thermal contact burns. Scald burns are the most common and typically occur from forced immersion in hot liquids, usually of the buttock, or in a stocking-and-glove distribution. Scald burns generally have sharp demarcation, uniform depth, and lack splash or drip marks that would be seen in an accidental immersion. Thermal burns occur from contact with hot objects, of which branding with metal implements or cigarettes is a common presentation. Concerning features of burns include:
Location on the hands (especially the dorsum), legs, feet, or buttocks
Patterned contact burns in the shape of an object (such as a fork, clothing iron, curling iron, or cigarette lighter)
Sharp stocking-and-glove pattern with sparing of the flexed protected areas (the classic forced immersion burn pattern)
There are various non-accidental fracture patterns, several with high specificity as described below:
Classic metaphyseal lesion (CML) – Also known as “bucket handle fractures” or “corner fractures,” these fractures are highly specific in children less than one year old. They result from a shearing force applied to a long bone, which causes avulsion of the metaphysis. These fractures are not associated with falls.
Multiple posterior and/or lateral rib fractures – These fractures also have a high correlation with NAT in children less than one year old. They arise from a specific mechanism – grasping the child around the torso and exerting a squeezing/compressive force. These fractures are more likely to affect the rib head and neck given the closer proximity to the transverse processes of the spine. NAT should especially be considered when healing fractures are found in a child without recent CPR.
Clavicular fractures and spiral fractures of long bones in nonambulatory children
Multiple fractures, especially if in different stages of healing
Spinous process fractures
Of note, spiral fractures of long bones generally result from twisting injuries (indicating NAT), but can occur accidentally from falls in ambulatory children. Therefore, these fractures (especially if coupled with clavicular fractures) are more specific for NAT in younger patients, and the specificity decreases with advancing age. Other described non-accidental patterns to consider include epiphyseal separations, vertebral body fractures and separations, digital fractures, linear and complex skull fractures, and subperiosteal bone formation. These patterns have low to moderate specificity for NAT. 
Abusive Head Trauma
Abusive head trauma (AHT) is the most fatal form of non-accidental injury in children. In fact, about 80% of deaths from NAT are caused by AHT and only 15% of patients with AHT survive without any sequelae.  AHT is a spectrum of injuries including collisions with stationary objects, direct blows to the head, and a repetitive acceleration- deceleration injury, also known as “Shaken Baby Syndrome.” Infants are particularly vulnerable to traumatic brain injury from shaking due to the relative weight of the head compared to the body, coupled with weak neck musculature.  If AHT is suspected, a non-contrast head CT should be obtained even with a nonfocal neurologic examination, because occult intracranial injury is common. Make sure to use age-appropriate dose reduction to minimize radiation exposure and if the CT scan is normal, consider further work-up with an MRI.
Although there are many ocular manifestations associated with non-accidental head injuries, retinal hemorrhages occur most often (about 60-85% of non-accidental head injuries).  Suspicion for NAT should be especially heightened when retinal hemorrhages are found in combination with signs of head trauma. Other ocular manifestations of NAT include periorbital hematoma, eyelid laceration, subconjunctival hemorrhage, subluxed or dislocated lens, cataracts, glaucoma, anterior chamber angle regression, iridiodialysis, retinal dialysis or detachment, intraocular hemorrhage, optic atrophy, or papilledema. 
Management and Disposition
All patients with suspected NAT should be admitted for protection and coordination of care even if they are clinically stable. Child Protective Services (CPS) must be notified, and engagement with the institutional social worker and child abuse team is recommended. It is important to note patients with NAT often have worse outcomes than other assault patients despite similar mechanisms of injury with intent to harm.  These patients often require close monitoring with Intensive Care Unit (ICU) resources. Patients with NAT should undergo a full skeletal survey as indicated in Figure 6 with additional imaging (CT, MRI) tailored to each patient. For instance, CT abdomen and pelvis should be obtained per general trauma guidelines, particularly if there is suspicion for solid organ or visceral injury.
Other diagnoses to consider in these patients include metabolic bone disease (such as rickets, Caffey disease, and osteogenesis imperfecta), blood dyscrasias, benign enlarged subarachnoid spaces (BESS), glutaric aciduria type 1 (which causes brain atrophy and subdural fluid collections).  However NAT is far more common than these diagnoses and carries significant morbidity and mortality when overlooked so should be considered and worked-up prior to these diagnoses.
Pediatric NAT causes significant morbidity and mortality, and therefore EPs must maintain a high degree of suspicion for this diagnosis.
Red flags during evaluation include a changing or inconsistent history, injuries inconsistent with the history, an unexpected mechanism of injury based on the child’s developmental status, and delayed presentation despite significant injury.
Risk factors for NAT include children younger than school age (with children younger than 1 being most vulnerable), family dysfunction, prematurity, multiple medical conditions, young/female parent, poor social support, unplanned or unwanted pregnancy, poor prenatal care, numerous separations from the child in the first year of life, and history of psychiatric issues, stressors, criminal activity, or childhood abuse or abandonment in the abuser.
Although physical exam findings can be non-existent or non-specific, highly specific findings include bruising over soft tissue areas; bruises/burns that are patterned take the form of an object; any bruising in a non-mobile child; scald burns on the hands, legs, feet, or buttocks; and stocking-and-glove patterned burns.
Highly concerning fracture patterns include classic metaphyseal lesions (“bucket handle fractures” or “corner fractures”), multiple posterior and/or lateral rib fractures, clavicular or spiral long bone fractures in any nonambulatory child, multiple fractures, fractures in different stages of healing, scapular fractures, sternal fractures, and spinous process fractures.
There is a wide range of ocular manifestations in NAT but the most common manifestation is retinal hemorrhage(s).
AHT carries the highest mortality rate of all the injuries associated with NAT. Any suspicion for AHT warrants consideration of a non-contrast head CT.
Notify Child Protective Services (CPS) and admit these children for further NAT work-up including a full skeletal survey.
Excellent overview of NAT in the Emergency Department with emphasis on risk factors and manifestations. I want to add a few pearls about NAT and then will focus my commentary on NAT management in the ED as well as discussion with families, as this was recently a large quality improvement project in our pediatric tertiary care center.
Neglect is the most common form of child abuse accounting for about two-thirds of all forms of abuse and often accompanies other forms of abuse. (1) Neglect is involved in about 50% of all cases of fatal child abuse. (1) Among children less than 1 year of age, 25% of fractures are a result of abuse. (2) Consider two things: does the explanation the provider stated account for the fracture the child has sustained? Is the child developmentally capable of the action being described? After 2 years of age, the history and physical exam should determine the imaging required. Over 5 years of age, the yield of unsuspected fractures from a skeletal survey is only 9%, making this group more amenable to selective radiographic studies. (3)
Diagnosis of NAT in children remains a challenge due to provider bias, preconceptions, and failure to recognize the presentation as possible abuse. (4,5) As a result, these injuries may go undetected, leading to further injury prior to diagnosis. An estimated 25% of children ultimately diagnosed with NAT have a sentinel injury prior to their abuse diagnosis. (6,7) Of abused children with a previous sentinel injury, the most common were a bruise (80%), a torn frenulum (11%), or a fracture (7%). (8) A large retrospective chart review estimated 80% of deaths from unrecognized abusive head trauma may have been prevented by earlier detection of NAT. (6) The American Academy of Pediatrics (AAP) states that “ANY injury to a young, pre-ambulatory infant” suggests abuse. (9)
At our institution, a team of pediatric emergency medicine physicians and child abuse pediatricians convened to develop and implement a standardized NAT guideline for providers in the ED when evaluating children with suspected NAT (Figure 1 Standardized Physical Abuse Guideline). This work stemmed from a chart review showing there was significant variability in the evaluation and management of children with concern for NAT in our Pediatric Emergency Department. The guideline was based on current peer reviewed literature as well as local expert consensus. It is divided into three separate age groups: < 6 months, 6-12 months, and >12-36 months. Age groups were determined based on risk of injury at different age levels in described literature, acquisition of milestones as age progresses, and increased ability for young children to show specific signs of injury with increasing age.
Lastly, the evaluation of NAT is stressful for both families and healthcare providers. The second page of our NAT guideline gives a sample script for EPs when discussing the non-accidental trauma evaluation for children. It states, “Any time a child comes to the hospital with this injury/these injuries, we evaluate for other injuries. Sometimes a child can have internal injuries such as fractures, head injury or abdominal injuries that we cannot see on the outside. Just like you, we want to make sure that your child is okay, so it is important to do this testing. We will also have our social worker come talk to you. This is a standard part of our evaluation. We are happy to answer any questions along the way”. It is important to acknowledge that this process is stressful, time consuming, and not comfortable for the child. Explaining each part of the process is important. Ensure that you use language that is non-accusatory. As EPs, we are not the ones to identify who the perpetrator is/was, but rather ensure the full NAT evaluation is completed and allow social work and/or Child Protective Services to determine further action.
Non-accidental trauma remains too prevalent in our country. Literature continues to show that unrecognized NAT leads to worse injuries and sometimes fatality. Continuing knowledge and education about injuries suspicious for NAT for EPs remains imperative. Standardized evaluations and real time order sets can increase appropriate management of NAT in the Emergency Department.
Dubowitz H. Epidemiology of Child Neglect. CAN 2011, pp 28-34.
Kaczor K, Clyde Pierce M. Abusive Fractures. CAN 2011, pp 275-295.
Martich KV. Imaging of Skeletal Trauma in Abused Children. CAN 2011, pp 296-308.
Higginbotham N, Lawson KA, Gettig K, et al. Utility of a child abuse screening guideline in an urban pediatric emergency department. J Trauma Acute Care Surg. 2014;76(3):871-877.
Tiyyagura GK, Gawel M, Koziel JR, et al. Barriers and facilitators to detecting child abuse and neglect in general emergency departments. Annals of Emergency Medicine. 2015;66(5):447-454.
Jenny C, Hymel K, Ritzen A, et al. Analysis of missed cases of abusive cases of head trauma. JAMA. 1999;282:621-6.
Rangel EL, Cook BS, Bennett BL, et al. Eliminating disparity in evaluation for abuse in infants with head injury: use of a screening guideline. Journal of Pediatric Surgery. 2009; 44(6):1229-34.
Sheets LK, et al. Injuries in Infants Evaluated for Child Physical Abuse. Pediatrics. 2013, pp 701-707.
Christian CW, Committee on Child Abuse and Neglect. The evaluation of suspected child physical abuse. Pediatrics. 2015;135:e1337–e1354.
Lauren C. Riney, DO
Division of Emergency Medicine
UC Department of Pediatrics
How to Cite this Post
[Peer-Reviewed, Web Publication] Loke D, Amick A. (2019, Oct 7). Non-Accidental Trauma. [NUEM Blog. Expert Commentary by Riney C]. Retrieved from http://www.nuemblog.com/blog/nonaccidental-trauma.
Other Posts You Might Enjoy
Pfeifer, C.M., Hammer, M.R., Mangona, K.L., & Booth, T.N. (2017). Non-accidental trauma: the role of radiology. Emerg Radiol, 24, 207-213.
Kim, P.T. & Falcone, R.A. (2017). Non-accidental trauma in pediatric surgery. Surgical Clinics of North America, 97.1, 21-33.
Child maltreatment 2014. Report, Children’s Bureau. Washington, DC: U.S. Department of Health and Human Services; 2014. Available at: http://www.acf. hhs.gov/sites/default/files/cb/cm2014
Paul, A.R. & Adamo, M.A. (2014). Non-accidental trauma in pediatric patients: a review of epidemiology, pathophysiology, diagnosis and treatment. Transl Pediatr, 3, 195-207.
Maguire, S., Mann, M.K., Sibert, J. & Kemp, A. (2005). Are there patterns of bruising in childhood which are diagnostic or suggestive of abuse? A systematic review. Arch Dis Child, 90, 182-186.
Boos, S.C. (2017). Physical child abuse: Recognition. Retrieved April 21, 2017, from http://www.uptodate.com
Hobbs, C.J. (1986). When are burns not accidental? Archives of Disease in Childhood, 61, 357-361.
Binenbaum G., Rogers, D.L., Forbes, B.J., Levin, A.V., Clark, S.A., Christian C.W., Liu, G.T., & Avery R. (2013). Patterns of retinal hemorrhage associated with increased intracranial pressure in children. Pediatrics, 132, 430-434.
Litz, C.N., Ciesla, D.J., Danielson, P.D. & Chandler, N.M. (2017). A closer look at non-accidental trauma: Caregiver assault compared to non-caregiver assault. Journal of Pediatric Surgery, 52, 625-627.