Non-Accidental Trauma


A hypothetical 7 month-old infant presents to the emergency department for mild respiratory distress. There is no recent illness or symptoms to explain the infant’s tachypnea and mild hypoxia. There is no visible bruising on exam. The parent states that the infant is starting to pull to stand but does not yet cruise. They have had several falls onto their tile kitchen floor. The CXR (below) is read by the radiologist left posterior rib fractures in ribs 4-8.

Case courtesy of Dr. George Harisis, From the case Non-Accidental Injury

Case courtesy of Dr. George Harisis, From the case Non-Accidental Injury

Highly Specific Fracture Patterns for Non-Accidental Trauma

A helpful adage: “Those that don’t cruise rarely bruise.” Approximately 80% of NAT occur in children less than 18 months old. A study evaluating bruising in normal infants demonstrated that only 0.01% (6 of 465) of pre-cruisers had ecchymoses. Soft tissue findings, like bruises, may tip you off to underlying or fractures either underlying or elsewhere. Pierce et al. (2010) highlighted the value of the TEN 4 FACES mnemonic to highlight concerning bruises that should prompt a workup for NAT: bruises to the Torso, Ears, and Neck in children < 4, any bruising on immobile children < 4 months, or bruising to the Frenulum, Angle of the mandible, Cheek, Eyelid, or Sclera. 

Fractures are the second most common finding in pediatric non-accidental trauma after bruising or other soft-tissue injuries. Although there are often similar fracture patterns seen in both accidental and non-accidental trauma, the fractures described below are the most highly specific for non-accidental trauma and should heighten your suspicion for intentional physical abuse:

Metaphyseal Fracture (aka “Corner” or “Bucket Handle” fractures)

A metaphyseal fracture is made up of microfractures perpendicular to the long axis of the bone, most commonly the distal ends of the tibia, femur, humerus. These microfractures are caused by the shearing forces of ligaments when a child who cannot control their limbs is shaken forcefully while held around their torso. It is highly specific fracture and is almost universally considered pathognomonic for non-accidental trauma.

Case courtesy of Dr. Basab Bhattacharya, From the case Non-Accidental Injuries

Case courtesy of Dr. Basab Bhattacharya, From the case Non-Accidental Injuries

Case courtesy of Dr. JR Dwek. From the The Radiographic Approach to Child Abuse.

Case courtesy of Dr. JR Dwek. From the The Radiographic Approach to Child Abuse.

Posterior Rib Fractures

Posterior rib fractures occur when enough anteroposterior chest compression is generated to cause movement of the posterior rib that acts as a lever over the transverse spinal process. This, like with metaphyseal fractures, can be caused by holding and shaking an infant with two hands around the ribcage. Other mechanisms include “marked forward decceleration into a solid object” in MVCs or in other non-accidental trauma. Biomechanically, the amount of force needed to cause leverage against the transverse process cannot be replicated when the patient is lying with their back against a flat surface, as is the case in CPR. A small post-mortem study of infants that had received even two-handed CPR supports this: they found anterolateral fractures but no occurrences of posterior rib fractures. In studies by Kleinman et al. (1997) and by Barsness et al. (2003) Posteromedial rib fractures have a very high positive predictive value (95%) for NAT and have the highest specificity for NAT.

Case courtesy of Dr. Paula Brill, From the case Non-Accidental Trauma.

Case courtesy of Dr. Paula Brill, From the case Non-Accidental Trauma.

The Three S’s: Spinous Process, Scapular, and Sternal Fractures

Though seen less often, spinous process, scapula, and sternum fractures round out the top most specific fractures for non-accidental trauma. Sternum and scapular fractures occur in the setting of a direct blow of unusual amounts force and are unexplained in the normal handling of most infants. As with other fractures, it is important to determine if the provided history matches the mechanism.

Other Specific Fracture Findings

  • Clavicular fractures (after the period explained by birth trauma)

  • Epiphyseal separations

  • Vertebral body fractures/separations

  • Digital fractures

  • Complex skull fractures

Next Steps

  • Mandatory reporting to Child Protective Services

  • Per the American Academy of Pediatrics, all patients undergoing workup for non-accidental trauma should be admitted to the hospital.

  • Complete a skeletal survey, which involves approximately 21 x-rays focusing on each individual limb or body part.

  • Order a CT brain to evaluate both the skull and underlying brain. Reconstructions of specific CTs may allow rotation of images and better identification of fractures.  

What Are We Missing?

In the most recent data published by the National Child Abuse and Neglect Data System, there were 1,585 fatalities due to child abuse and neglect in 2015. Approximately 44% percent of those suffered death due to physical abuse and almost 75% were children <3 years old.

A small study comparing known instances of child abuse fatalities with local medical records found that 30% of children who subsequently died from non-accidental trauma had interactions with health care for reasons other than well-child checks. Nearly 20% of those visits occurred within one month of their death. Albeit brief, emergency department visits may be the only interaction these children have with the health care system represent a critical opportunity for intervention.


Faculty Reviewer: Dr. Adam Aluisio


  1. Baldwin K, Pandya NK, Wolfgruber BA, et al. Femur Fractures in Pediatric Population: Abuse or Accidental Trauma? Clin Orthop Relat Res. 2011 Mar; 469(3):798-804.

  2. Barsness KA, Cha ES, Bensard DD, Calkins CM, Partrick DA, et al. The positive predictive value of rib fractures as an indicator of nonaccidental trauma in children. J Trauma. 2003;54:1107–1110.

  3. Bechtel K. Physical Abuse of Children: Epidemiology of Child Abuse in the United States. Emergency Medicine Reports. 2003 Mar.

  4.  Child Welfare Information Gateway. (2017). Child abuse and neglect fatalities 2015: Statistics and interventions. Washington, DC: U.S. Department of Health and Human Services, Children’s Bureau.

  5. Christian CW. Committee on Child Abuse and Neglect. The Evaluation of Suspected Child Physical Abuse. Pediatrics. 2015; 135(5):1337-1354. 

  6. Dwek JR. The Radiographic Approach to Child Abuse. Clinical Orthopaedics and Related Research. 2011;469(3):776-789.

  7. King WK, Kiesel EL, Simon HK. Child Abuse Fatalities: Are We Missing Opportunities for Intervention? Pediatric Emerg Care. 2006;22(4):211-214.

  8. Kleinman PK, Perez-Rossello JM, Newton AW, et al. Prevalence of the classic metaphyseal lesion in infants at low versus high risk for abuse. AJR Am J Roentgenol. 2011 Oct;197(4):1005-8.

  9. Kleinman PK, Schlesinger AE. Mechanical factors associated with posterior rib fractures: laboratory and case studies. Pediatr Radiol. 1997(27): 87-91.

  10.  Leaman LA, Hennrikus WL, Bresnahan JJ. Identifying non-accidental fractures in children aged <2 years . Journal of Children’s Orthopaedics. 2016;10(4):335-341.

  11. Matshes EW, Lew EO. Two-Handed Cardiopulmonary Resuscitation Can Cause Rib Fractures In Infants. Amer Journal of Forensic Med and Pathology. 2010 Dec; 31(4): 303-307.

  12. Paddock M, Sprigg A, Offiah AC. Imaging and reporting considerations for suspected physical abuse (non-accidental injury) in infants and young children. Part 1: initial considerations and appendicular skeleton. Clinical Radiology. 2017 Mar;72(3):179-188

  13. Pierce MC, Kaczor K, Aldridge S, O'Flynn J, Lorenza DJ. Bruising Characteristics Discriminating Physical Child Abuse From Accidental Trauma. Pediatrics. 2019;125(1):67-74.

  14. Sugar NF, Taylor JA, Feldman KW, and the Puget Sound Pediatric Research Network. Bruises in Infants and Toddlers Those Who Don't Cruise Rarely Bruise. Arch Pediatr Adolesc Med. 1999;153(4):399–403.

Seatbelt Sign of the Neck in Pediatric Trauma


The patient is a 11 y.o. male with no past medical history who presents as trauma activation after MVC. Patient was the restrained front seat passenger in a head on collision. Denies LOC. Per EMS, patient was found in police cruiser on arrival, patient states he walked at scene. +Seatbelt sign. Vitals stable. Patient states his pain is worst in his neck, rates pain 6/10. Denies numbness/tingling.

 Vital Signs:  

BP 116/47  | Pulse (!) 115  | Resp 19  | Wt 37.5 kg  | SpO2 100%

Pertinent physical exam:

AIRWAY: intact  

BREATHING: non-labored Breath sounds: Clear to auscultation bilaterally

CIRCULATION: pulse palpable: Bilateral Radial, DP and PT pulses are normal and symmetric

Capillary refill: normal; less than 2 seconds

Head: normocephalic / atraumatic no hematoma, no abrasions

ENT: tympanic membranes bilaterally clear

Neck: trachea appears midline, there is a cervical collar in place. Abrasion/seatbelt sign over R neck. +midline c-spine ttp without stepoffs

Respiratory: clear to auscultation bilaterally

Cardiovascular: regular rate and rhythm. Seat belt sign across L anterior chest

Rectal: normal tone

Abdomen: soft, non-distended, normal bowel sounds, nonperitoneal with tenderness to palpation in RLQ. Seatbelt sign across lower abdomen and L anterior thigh

Back: no tenderness to palpation in t-spine or l-spine. No step offs. No abrasions or bruises.

Pelvis: non-tender, stable to anterior-posterior/lateral compression

Genitourinary: normal genitalia

Musculoskeletal: no palpable long bone deformities, no bony tenderness to palpation

Skin: grossly intact

Neurologic: GCS: eyes 4, best verbal response 5, best motor response 6 TOTAL GCS SCORE: 15. Bilateral upper extremity strength 5/5 at deltoids, biceps, triceps, and handgrip. Bilateral lower strength at 5/5 at hip flexion, dorsi-/plantarflexion.

Should this patient receive a CTA of the neck?


Review of the literature 

Blunt cervical vascular injury

Blunt cervical vascular injury (BCVI) has an incidence of 0.03-0.9% in pediatric blunt trauma. BCVI may cause ischemia and other neurologic sequelae. Most BCVI are treated medically with aspirin or anticoagulation. Higher grade lesions may require intervention including endovascular stenting or ligation. The development of focal neurologic findings may be delayed up to 10-72 hours, complicating diagnosis in the acute setting.[1]

Currently, the Eastern Association for the Surgery of Trauma (EAST) recommends that pediatric patients should be screened by adult criteria, called the Denver or Memphis criteria. The Denver criteria include “focal neurologic deficit, arterial hemorrhage, cervical bruit/thrill (<50 yo), infarct on head CT, expanding neck hematoma, neuro exam inconsistent with head CT, midface fractures, c-spine injuries, basilar skull fractures, GCS <8, hanging with anoxic brain injury, seat belt abrasion or other soft tissue injury of the anterior neck resulting in significant swelling or altered mental status. Isolated seatbelt sign without other neurologic symptoms has not been identified as a risk factor.”[2]

Evaluation with computed tomography angiogram (CTA) of the brain and neck represents a dose of radiation 8 times higher (16.4 millisieverts) than compared to a noncontrast CT of the brain and neck (2 millisieverts).[1] Exposure to ionizing radiation in the pediatric population has been shown to increase risks of cancers, especially leukemia, breast, and thyroid cancer.[3]

A recent, large study identifying risk factors associated with blunt cervical vascular injury (BCVI) examined 11,446 pediatric blunt trauma patients, with 375 (3.3%) undergoing CTA imaging. Fifty-three patients (0.4%) had cerebrovascular injuries, representing 0.5% of all pediatric blunt trauma patients and 14% of all blunt trauma patients screened with CTA.[1]

They found a seatbelt sign on the neck did not predict vascular injury.[1] These findings are consistent with other studies which did not find an association between  a cervical seatbelt sign and BCVI.[2,4] Furthermore, they identified independent predictors of cervical vascular injury: presence of cerebral hemorrhage, infarct on head imaging, cervical spine fracture, and basilar skull fracture.[1] Other studies have found associations between BCVI and clavicular fractures,[5] fracture through the carotid canal,[6] petrous temporal bone fracture,[6] GCS < 8,[6] focal neurological deficit,[6] and stroke on initial CT.[6]


Case Resolution

Due to the patient’s c-spine tenderness, in addition to the presence of a seat belt sign, the patient underwent imaging with a CT c-spine and a CTA brain and neck. He also had a CT of his abdomen.

All of his imaging (CTA brain/neck, CT c-spine, and CT abdomen and pelvis were all negative. The patient was admitted to the surgical service for pain management and serial exams of his abdomen. His repeat exam was normal and patient was discharged the following day.

Based on the above guidelines, a CTA of the brain and neck was not indicated.

Children of a certain size and age should not be sitting in the front seat and need booster seats. Here’s more information on that:


Faculty Reviewer: Dr. Jane Preotle


  1. Irma T. Ugalde IT,  MD, Claibrne MK, Cardenas-Turanzas M, Shah MN, Langabeer JR, Patel R. Risk Factors in Pediatric Blunt Cervical Vascular Injury and Significance of Seatbelt Sign. West J Emerg Med. 2018 Nov; 19(6): 961–969.

  2. Denver screening criteria. WikEM

  3. Risk of Ionizing Radiation Exposure to Children: A Subject Review. Committee on Environmental Health, American Association of Pediatrics. Pediatrics. 1998; 101(4).

  4. Desai NK, Kang J, Chokshi FS. Screening CT Angiography for Pediatric Blunt Cerebrovascular Injury with Emphasis on the Cervical “Seatbelt Sign” American Journal of Neuroradiology September 2014, 35 (9) 1836-1840.

  5. Lew SM, Frumiento C, Wald SL Pediatric blunt carotid inury: a review of the National Pediatric Trauma Registry. Pediatr Neurosurg, 1999; 30(5): 239-44. 

  6. Ravindra VM, Bollo RJ, Sivakumar W, Akbari H, Naftel RP, Limbrick DD, JEa A, Gannon S, Shannon C, Birkas Y, Yang GL, Prather CT, Kestle JR, Riva-Cambrin J. Predicting Blunt Cerebrovascular Injury in Pediatric Trauma: Validation fo the “Utah Score.” J Neurotrauma. 2017; 34(2): 391-399.

Pediatric Baclofen Overdose


A 16 year-old male presented to the emergency department after intentional overdose of 200 mg of baclofen. The patient was found in his bedroom by family members approximately 10 hours after ingestion with reported twitching, vomiting, unresponsiveness, and possible seizure activity. On arrival to the emergency department, the patient was awake, alert, and oriented; tearful, but otherwise asymptomatic.


Physical Exam

Vital signs included blood pressure 150/76, heart rate 112, temperature 99.2 F (37.3 C), respiratory rate 26, SpO2 97%. Physical exam showed frequent bilateral upper greater than lower extremity myoclonic jerks. The patient’s neurologic exam and physical exam were otherwise unremarkable.



The patient’s toxicological workup was pertinent for undetectable salicylate, acetaminophen, and ethanol levels. His urine drug screen was negative. EKG showed sinus tachycardia with normal QRS and QTc intervals. A head CT previously completed at an outside hospital was normal. The rest of his laboratory evaluation including CBC, electrolyte levels, and hepatic function tests, was unremarkable.



Baclofen is a centrally-acting skeletal muscle relaxant which functions as a GABA-B receptor agonist, believed to inhibit synaptic transmission of signals to the muscle from the spinal cord. Baclofen is typically prescribed for symptoms related to severe muscle spasm, such as in spinal cord injury or chronic neurologic disease (e.g. multiple sclerosis.) The medication can be administered either orally, or via an intrathecal pump.

Baclofen is absorbed rapidly from the GI tract in a dose-dependent manner, with peak serum levels occurring after ~1 hour (with a range of 0.5-4 hours). However, this has been found to be highly variable in pediatric patients. Its volume of distribution is also highly variable among pediatric patients, with nearly 50% interindividual variability. The reason for this variability is not clearly understood.

Baclofen is primarily excreted renally, with a serum half-life of 4.5 hours in pediatric patients, and a CSF half-life of 1.5 hours in intrathecal administration.

Baclofen overdose typically manifests with neurologic symptoms and dysautonomia. Symptoms are variable and may include: CNS depression/coma, hypotonia, hyporeflexia, seizure, respiratory depression, tachycardia, bradycardia, hypertension, hypotension, and arrhythmia. In adults, ingestions of >200 mg appear to correlate with an increased risk of respiratory failure, mechanical ventilation requirement, and ICU admission time.

 In an unfortunate case series, a group of teenagers overdosed on 60-600 mg of baclofen for recreational purposes. Patients displayed symptoms of overdose in 1-2 hours. 9 of 14 were intubated. Of 8 that were followed longitudinally: 7 were comatose, 6 were hypothermic, 5 were bradycardic, 4 were hypertensive, 8 were hyporeflexic, 3 had PVCs, and 2 had tonic-clonic seizures. The mean time of intubation was 40 hours. All patients recovered.



Treatment of baclofen overdose is primarily supportive with IV fluids, hemodynamic and respiratory support, and antiepileptics as needed. Activated charcoal may also be considered. There are case reports of physostigmine being effective in low/moderate overdoses; however, its use is controversial. Finally, hemodialysis does shorten clearance time and resolution of toxicity in patients with normal and impaired renal function.



Patients should be monitored until symptoms resolve. Depending on the dose ingested, the length of time since ingestion, clinical status, and laboratory analysis, they may be monitored either on the floor with telemetry or in the ICU.



The patient was treated with a 1 liter normal saline bolus and 1 mg of lorazepam IV; his hemodynamics normalized, his myoclonic twitching resolved, and he remained stable without seizure or complication throughout emergency department stay. He was admitted to the pediatric ICU for overnight monitoring, where he remained asymptomatic and had normal vital signs. The patient was then transferred to inpatient psychiatry for mental health treatment.



Baclofen overdoses typically present with a combination of altered mental status and or seizure, hypotonia, and dysautonomia. Treatment is primarily supportive and most patients recover with IV fluids, hemodynamic and respiratory support, and antiepileptics as-needed (usually benzodiazepines.) Activated charcoal should be considered. Patients should be admitted either to medical floor with telemetry or ICU depending on clinical status; they should remain admitted until symptoms resolve (typically several days, depending on ingested dose).

Faculty Reviewer: Dr. Jane Preotle