Residency · Residency · Pediatrics

Recognizing Non-Accidental Trauma

Introduction

Non-accidental trauma (NAT), or child physical abuse, is a critical diagnosis that pediatricians must be prepared to identify. Approximately 680,000 children are confirmed victims of abuse or neglect annually in the United States, and 1,750 children die each year from maltreatment. The majority of fatalities occur in children under 3 years of age. The challenge lies in distinguishing inflicted injuries from accidental ones, as abused children often present to medical care with explanations that may initially seem plausible. A high index of suspicion, systematic evaluation, and understanding of developmental milestones are essential to protecting vulnerable children.

Epidemiology and Risk Factors

Epidemiology

Infants and toddlers are at highest risk for severe and fatal abuse, with a peak age under 1 year. Abusive head trauma is the leading cause of death from physical abuse. Boys and girls are equally affected in early childhood. Perpetrators are most often parents or caregivers, accounting for 80-90% of cases.

Risk Factors

Child factors include prematurity, disability, chronic illness, colic, twins or multiples, and unwanted pregnancy. Caregiver factors include young parental age, substance abuse, mental illness, domestic violence, history of being abused, and social isolation. Environmental factors include poverty, unemployment, unstable housing, and lack of social support. These are associations, not causation, and abuse occurs across all socioeconomic, racial, and ethnic groups.

Sentinel Injuries

Sentinel injuries are seemingly minor injuries (bruises, oral injuries) in pre-cruising infants that may precede more severe abuse. Bruising in a non-mobile infant ("those who don't cruise rarely bruise") is highly suspicious. The TEN-4 rule identifies high-risk bruising: bruises to the torso, ears, or neck in a child under 4 years, or any bruise in an infant under 4 months, should raise concern for abuse. Studies show that sentinel injuries are present in 25-30% of cases before a severe abusive event is recognized.

<image>Body diagram of an infant and toddler highlighting the distribution of bruising patterns: typical accidental bruising locations (shins, forehead, bony prominences in mobile children) versus concerning locations for non-accidental trauma (torso, ears, neck, cheeks, buttocks, genitalia, and any bruising in a pre-mobile infant), with the TEN-4 rule criteria annotated</image>

Patterns of Injury Suspicious for Abuse

Skin and Soft Tissue Injuries

Bruises are the most common manifestation of physical abuse. Concerning patterns include patterned bruises (loop marks from cords, belt buckle outlines, hand prints, bite marks), bruises in various stages of healing (though dating bruises by color is unreliable), and bruises in unusual locations (trunk, face, ears, neck, buttocks). Burns suspicious for abuse include immersion burns with clear demarcation lines (stocking-and-glove distribution), contact burns with identifiable shapes (cigarette, iron), and bilateral symmetric involvement. Bite marks with an inter-canine distance greater than 3 cm suggest an adult bite, and forensic odontology consultation is recommended.

Fractures

Fractures with the highest specificity for abuse include classic metaphyseal lesions (corner or bucket-handle fractures), posterior rib fractures, scapular fractures, spinous process fractures, and sternal fractures. Multiple fractures at different healing stages, bilateral fractures, and complex skull fractures also carry high specificity. Any fracture in a non-ambulatory infant requires investigation for NAT. Rib fractures in infants, especially posterior, result from anteroposterior compression during violent shaking or squeezing and are rarely caused by CPR in infants.

Abusive Head Trauma (AHT)

Previously termed "shaken baby syndrome," abusive head trauma is now a broader term encompassing multiple mechanisms. The classic triad (though not required for diagnosis) includes subdural hemorrhage, retinal hemorrhages, and encephalopathy. Subdural hematomas are often bilateral, interhemispheric, or posterior fossa and may be of different ages. Retinal hemorrhages are present in 85% of AHT cases and are typically extensive, multilayered, and extending to the periphery, which is rare in accidental trauma or medical conditions. Additional findings include bridging vein tears, diffuse axonal injury, and cerebral edema. Ophthalmology consultation for dilated fundoscopic examination should be obtained in all suspected cases.

Abdominal Injuries

Abdominal injuries are the second leading cause of death from child abuse. They include duodenal hematoma, pancreatic injuries, liver laceration, mesenteric tears, and hollow viscus perforation. These often present with nonspecific symptoms (vomiting, abdominal distension, irritability). Delay in seeking care is a red flag, and abdominal injuries from abuse carry higher mortality than accidental abdominal trauma.

<image>Radiographic illustration panel showing characteristic skeletal findings in non-accidental trauma: classic metaphyseal lesion (corner fracture and bucket-handle fracture) of the distal tibia, posterior rib fractures in an infant, healing fractures of different ages in a long bone, and a complex skull fracture with associated subdural hematoma on CT imaging</image>

Red Flags in the History

An inconsistent history where the mechanism of injury does not match the pattern or severity of findings is a major red flag. Other warning signs include a changing or evolving story with different explanations provided to different providers, delay in seeking medical care disproportionate to the severity of injury, developmental incompatibility where the attributed mechanism is beyond the child's developmental capability, a history of an "accident-prone" child or frequent ED visits for injuries, blaming siblings or other children for injuries in a pre-verbal child, and concerning caregiver behavior such as being overly calm or dismissive, avoiding eye contact, leaving before the evaluation is complete, or hostility toward staff.

Evaluation Protocol

Medical Workup

A skeletal survey is mandatory for children under 2 years with suspected abuse, including AP views of all long bones, hands, feet, spine, pelvis, and skull, with a follow-up survey in 2 weeks (which increases fracture detection by 25%). Head CT without contrast is obtained for all children under 6 months with any concern for abuse and for any child with neurologic symptoms. Abdominal labs (AST, ALT, lipase, amylase, urinalysis) screen for occult abdominal injury, with AST or ALT greater than 80 warranting CT abdomen. Ophthalmologic examination with dilated fundoscopy by an ophthalmologist is essential. Coagulation studies (PT, PTT, fibrinogen, von Willebrand panel) rule out bleeding disorders. Clinical photography of all injuries with a measurement scale should be obtained.

Differential Diagnosis

The differential includes bleeding disorders (hemophilia, von Willebrand disease, factor deficiencies, ITP), metabolic bone disease (osteogenesis imperfecta with blue sclerae, family history, and Wormian bones; rickets; Menkes kinky hair syndrome), skin conditions (Mongolian spots/dermal melanocytosis, phytophotodermatitis, folk remedies such as coining and cupping), and medical conditions mimicking AHT (glutaric aciduria type 1, benign enlargement of subarachnoid spaces, birth-related subdural hemorrhages). A thorough evaluation should seek to confirm or exclude these conditions.

<image>Clinical workflow algorithm for evaluating suspected non-accidental trauma in a child under 2 years, starting from clinical suspicion through history and physical examination, laboratory workup (coagulation studies, liver enzymes), skeletal survey, head CT, ophthalmologic exam, photography, social work involvement, and mandatory reporting steps</image>

Clinical Pearls

"Those who don't cruise rarely bruise" -- any bruising in a pre-mobile infant should prompt evaluation for abuse. The TEN-4 rule (torso, ears, neck bruising in children under 4 years; any bruise in infants under 4 months) identifies high-risk bruising patterns. Classic metaphyseal lesions and posterior rib fractures in infants are highly specific for inflicted injury. Retinal hemorrhages in abusive head trauma are typically extensive and multilayered; limited retinal hemorrhages may occur with accidental trauma but rarely extend to the periphery. A normal initial skeletal survey does not exclude abuse, and follow-up imaging at 2 weeks detects an additional 25% of fractures. Occult abdominal injury should always be screened for with AST, ALT, lipase, and urinalysis, as abdominal trauma from abuse often presents late.

References

1. Christian CW, Committee on Child Abuse and Neglect. The evaluation of suspected child physical abuse. Pediatrics. 2015;135(5):e1337-e1354.
2. Pierce MC, Kaczor K, Aldridge S, et al. Bruising characteristics discriminating physical child abuse from accidental trauma. Pediatrics. 2010;125(1):67-74.
3. Duhaime AC, Christian CW, Rorke LB, Zimmerman RA. Nonaccidental head injury in infants — the "shaken-baby syndrome." New England Journal of Medicine. 1998;338(25):1822-1829.
4. Kemp AM, Dunstan F, Harrison S, et al. Patterns of skeletal fractures in child abuse: systematic review. BMJ. 2008;337:a1518.

Read this lecture as Markdown