# Child Abuse and Non-Accidental Trauma Recognition
## Epidemiology
### Overview
Approximately 3.5 million child abuse reports are filed annually in the United States, with approximately 700,000 confirmed victims. Child abuse fatalities claim approximately 1,750 lives per year, with the majority of victims under 3 years old. Physical abuse accounts for 18 percent of confirmed cases, while neglect represents 75 percent. Emergency department physicians are often the first medical professionals to evaluate these children, and missed abuse has devastating consequences, with re-injury rates of 30 to 50 percent when abuse is not recognized on the initial encounter.
### Risk Factors for Abuse
Child factors that increase risk include age under 3 years (the highest-risk group), prematurity, disability, chronic illness, colic, and multiple births. Caregiver factors include substance abuse, mental illness, domestic violence in the home, social isolation, young or single parenting, and the caregiver's own history of being abused. Situational factors include poverty, unemployment, housing instability, and inadequate support systems.
## Injury Patterns Suspicious for Abuse
### General Red Flags
Several patterns should raise concern for abuse: a history that is inconsistent with the injury or with the child's developmental stage, changing or contradictory histories between caregivers, a delay in seeking medical care, injuries in pre-ambulatory children (those who cannot roll, crawl, or walk), pattern injuries such as loop marks, belt marks, bite marks, and circular cigarette burns, injuries in unusual locations (ears, neck, buttocks, genitalia, trunk), multiple injuries in different stages of healing, and a caregiver who appears unconcerned about the injury or hostile toward questioning.
### Bruising
Normal accidental bruising overlies bony prominences such as the shins, knees, forehead, and elbows. It increases with the child's mobility and is rare in pre-ambulatory infants. The clinical axiom is: "those who don't cruise rarely bruise."
Suspicious bruising patterns include any bruising in pre-ambulatory infants under 6 months and the TEN-4 rule, which identifies bruises on the torso, ears, or neck in children under 4 years, or any bruise in a child under 4 months, as highly suspicious for abuse. Patterned bruises such as linear marks from a belt or cord, loop marks from a cord or wire, hand prints, and bite marks are concerning, as are bilateral bruises and bruises appearing in clusters.
Cultural practices such as cupping (hijama), coining (cao gio), and moxibustion can produce skin findings that mimic abuse. Cultural competency is essential in these evaluations, but the presence of cultural marks should not automatically preclude abuse evaluation.
It is important to recognize that dating bruises by color is unreliable, and clinicians should not attempt to determine the age of a bruise based on its appearance.
### Fractures
Certain fracture patterns are highly specific for abuse. Classic metaphyseal lesions (corner or bucket-handle fractures) are caused by twisting or pulling forces on the extremity and are highly specific for abuse in infants. Posterior rib fractures result from squeezing or compression and are very specific for abuse. Fractures of the scapula, spinous processes, and sternum are uncommon locations for accidental injury and are therefore suspicious.
Moderately suspicious patterns include multiple fractures in different stages of healing, complex skull fractures (bilateral, depressed, or crossing suture lines), and femur fractures in a non-ambulatory child.
Common accidental fractures include clavicle fractures, distal radius fractures from falls on an outstretched hand (FOOSH), toddler's fractures (spiral tibial fractures in ambulatory toddlers), and linear parietal skull fractures with a clear short fall history.
A skeletal survey is indicated for all children under 2 years with suspected abuse.
### Burns
Suspicious burn patterns include immersion burns with a clear line of demarcation (the "waterline"), stocking or glove distribution on hands and feet, sparing of flexion creases (indicating the child was held in a flexed position), and buttock or perineal burns from forced sitting in hot water. Contact burns have a clearly defined shape matching the object used (iron, cigarette, curling iron). Bilateral symmetric burns are also unlikely to be accidental. By contrast, accidental burns typically have irregular borders, a splash pattern, asymmetric distribution, and occur on anterior surfaces.
### Abusive Head Trauma (AHT)
Formerly termed "shaken baby syndrome," the terminology has evolved to abusive head trauma. The mechanism involves acceleration-deceleration forces from shaking, impact, or both. The classic triad consists of subdural hematomas, retinal hemorrhages, and encephalopathy. Presentation includes altered mental status, seizures, apnea, vomiting, irritability, a bulging fontanelle, and increasing head circumference, but it may also present with non-specific symptoms such as fussiness, poor feeding, and lethargy. Retinal hemorrhages are present in 80 to 90 percent of cases, and multilayered hemorrhages that are too numerous to count and extend to the periphery are highly specific. The differential includes accidental head trauma, birth-related subdural hemorrhages (which resolve by 3 months), coagulopathies, and metabolic bone diseases such as osteogenesis imperfecta and rickets.
### Abdominal/Visceral Injuries
Abdominal injuries are the second leading cause of child abuse fatalities after head injuries. They are often occult and may present with vomiting, abdominal distension, or shock. Common injuries include liver laceration, splenic rupture, duodenal hematoma, pancreatic injury, and hollow viscus perforation. Screening with AST and ALT is recommended, and levels above 80 U/L are sensitive for hepatic injury and should prompt CT of the abdomen. Bruising of the abdomen or trunk may be absent despite significant internal injury.
## Evaluation Protocol
### History
A detailed history should be obtained from all caregivers separately. Documentation should include who was with the child, what happened (mechanism), when it happened (timing), when care was sought, and the child's developmental milestones (specifically whether the child is capable of the actions described). The history should be documented in the caregiver's own words using direct quotes.
### Physical Examination
A complete undressed examination should be performed, documenting all findings including normal examination areas. All injuries should be measured and diagrammed. Photographs should be taken with a ruler or measuring scale for reference. The examination should include the scalp, behind the ears, the frenulum (a torn frenulum may indicate forced bottle-feeding), the oral mucosa, trunk, buttocks, genitalia, and extremities. A dilated fundoscopic examination by ophthalmology should be performed when abusive head trauma is a concern.
### Laboratory Workup
The laboratory evaluation includes a CBC with differential, a comprehensive metabolic panel (with particular attention to AST and ALT as screening for abdominal injury), lipase and amylase, coagulation studies (PT, PTT, fibrinogen) to rule out a coagulopathy, urinalysis (hematuria suggests renal injury), and a urine drug screen if exposure is suspected. When bruising is the primary concern and a bleeding disorder needs to be excluded, a von Willebrand panel and factor levels should be considered.
### Imaging
A skeletal survey is mandatory for all children under 2 years with suspected abuse. It includes AP views of all long bones, AP and lateral views of the chest, AP view of the pelvis, AP and lateral views of the skull, and AP and lateral views of the spine. A follow-up skeletal survey at 2 weeks detects healing fractures that were not initially visible and increases the yield by 25 percent. For children aged 2 to 5 years, a selective skeletal survey is obtained based on clinical judgment. Children over 5 years generally do not need a skeletal survey because they can verbalize the locations of their pain.
Head CT should be obtained for any child with suspected abusive head trauma, altered mental status, or concerning neurologic findings. MRI of the brain is more sensitive than CT for diffuse axonal injury, parenchymal injury, and dating of subdural collections, and should be obtained when CT is positive or clinical suspicion is high. Abdominal CT should be obtained when AST or ALT exceeds 80 U/L or when there is clinical concern for abdominal injury.
## Mimics of Abuse
### Medical Conditions
Several medical conditions can mimic findings of abuse. Osteogenesis imperfecta causes multiple fractures and may present with blue sclerae, hearing loss, and a family history, though genetic testing may be needed. Rickets and vitamin D deficiency cause metaphyseal changes, but the pattern differs from classic metaphyseal lesions. Bleeding disorders including von Willebrand disease, hemophilia, factor deficiencies, and idiopathic thrombocytopenic purpura can cause unexplained bruising. Mongolian spots (dermal melanocytosis) are blue-gray birthmarks on the buttocks and back that can be mistaken for bruises but do not blanch and are present from birth. Ehlers-Danlos syndrome causes easy bruising with joint hypermobility. Henoch-Schonlein purpura produces palpable purpura on the buttocks and lower extremities. Various coagulopathies including factor deficiencies, liver disease, and DIC should also be considered.
### Cultural Practices
Cupping (hijama) leaves circular ecchymoses from suction cups. Coining (cao gio), a Vietnamese and Southeast Asian practice, produces linear ecchymoses from scraping with a coin. Moxibustion, used in Traditional Chinese Medicine, creates circular burns from heated herbs. These practices are not abuse in themselves, but the presence of cultural marks does not exclude concurrent abuse, and each case should be evaluated individually.
## Documentation and Reporting
### Documentation Standards
Documentation should use exact quotes from caregivers, record developmental milestones, describe injuries precisely (including location, size, shape, color, and pattern), include diagrams and photographs, and document the names of all individuals present during the visit. Personal opinions about guilt should not be included; only objective findings should be documented.
### Mandatory Reporting
All 50 states and all Canadian provinces have mandatory reporting laws. Reports should be made based on reasonable suspicion, and certainty is not required. Reports go to child protective services and local law enforcement as required by the jurisdiction. The reporter is protected from liability if the report is made in good faith, and failure to report is a legal violation that can result in criminal charges against the physician. The physician's role is to report and document; investigation is the responsibility of CPS and law enforcement.
### Disposition
If it is safe to discharge, CPS involvement should be arranged along with a safe caregiver assessment and close follow-up. If safety cannot be ensured, the child should be admitted for medical or social reasons, with a temporary custody order requested if needed. Emergency departments can place a medical hold on a child if immediate danger is suspected.
A body diagram of a child showing the distribution of accidental versus suspicious bruising. The left figure shows a "normal" accidental bruising pattern with bruises highlighted on bony prominences: forehead, shins, knees, and elbows. The right figure shows suspicious bruising patterns with highlighted areas: ears (TEN-4 rule), neck, trunk, buttocks, genitalia, and upper arms. Annotations emphasize the TEN-4 rule (Torso, Ears, Neck in children under 4 years; any bruise in children under 4 months). Pattern injuries are illustrated as insets: loop marks (from cords), belt marks (linear), cigarette burns (circular), and bite marks (paired crescentic).
A radiographic panel showing fractures associated with non-accidental trauma. Panel 1: Classic metaphyseal lesion (corner fracture) of the distal tibia in an infant — showing the small triangular fragment at the metaphysis (arrow). Panel 2: Posterior rib fractures on a skeletal survey — showing healing fractures at the costovertebral junctions bilaterally (arrows). Panel 3: Complex skull fracture — showing bilateral parietal fractures crossing the midline suture. Panel 4: Follow-up skeletal survey at 2 weeks showing periosteal reaction and callus formation at previously occult fracture sites (arrows), demonstrating the value of the repeat survey.
An illustration of immersion burn patterns in child abuse. Panel 1: Stocking/glove distribution — clear line of demarcation on the lower legs and feet, with uniform depth of burn below the waterline and sparing above. Panel 2: Buttock/perineal immersion burn — symmetric burns on the buttocks with sparing of the skin folds (flexion creases), indicating the child was held in a flexed position in hot water. Panel 3: An accidental scald burn for comparison — irregular borders, splash pattern, asymmetric distribution on the anterior surface. Key distinguishing features are labeled for each pattern.
## Clinical Pearls
"Those who don't cruise rarely bruise" — any bruise in a pre-ambulatory infant warrants an abuse evaluation. The TEN-4 rule identifies bruises on the torso, ears, or neck in children under 4 years, or any bruise in infants under 4 months, as suspicious for abuse. Classic metaphyseal lesions and posterior rib fractures are highly specific for abuse and rarely occur from accidental mechanisms. Dating bruises by color is unreliable, and clinicians should not attempt to determine the age of a bruise based on appearance alone. Retinal hemorrhages that are multilayered, too numerous to count, and extending to the periphery are strongly associated with abusive head trauma, and ophthalmology consultation should be requested. AST or ALT above 80 U/L in a child with suspected abuse should prompt CT of the abdomen, as occult abdominal injury is the second leading cause of abuse-related death. A follow-up skeletal survey at 2 weeks increases fracture detection by 25 percent and should always be arranged for children under 2 years. Suspected abuse should be reported based on reasonable suspicion because certainty is not required, investigation is not the physician's responsibility, and failure to report is a legal violation.
## References
- Christian CW, et al. The evaluation of suspected child physical abuse (AAP Clinical Report). *Pediatrics*. 2015;135:e1337-e1354.
- Pierce MC, et al. Bruising characteristics discriminating physical child abuse from accidental trauma (TEN-4 rule). *Pediatrics*. 2010;125:67-74.
- Maguire S, et al. Which clinical features distinguish inflicted from non-inflicted brain injury? A systematic review. *Arch Dis Child*. 2009;94:860-867.
- Kleinman PK. *Diagnostic Imaging of Child Abuse*. 3rd ed. Cambridge University Press; 2015.
- Lindberg DM, et al. Hepatic transaminase screening for abusive abdominal trauma. *Pediatrics*. 2013;131:e1013-e1018.