Evaluating children who present with fractures for abuse can be challenging but is important since fractures are one of the commonest presentations of physical injury in children who are physically abused. Understanding the biomechanics of various fracture morphologies, recognizing fracture patterns correlated with abuse, and utilizing evidence-based diagnostic imaging modalities and protocols supports the clinician’s ability to make a plan for further evaluation, make an informed assessment of the likelihood of abuse, and promptly contact child protection agencies.
Key points
- •
Clinicians caring for children should be competent in identifying skeletal injury patterns suggestive of abuse versus accidents to determine the need for further evaluation and safety assessment.
- •
Skeletal survey is an important radiologic tool in children less than 24 months to detect occult fractures in suspected victims of child abuse.
- •
Physical abuse is suspected when the purported mechanism of injury contradicts the biomechanics of the fracture pattern observed.
Introduction
Epidemiology
According to the 2022 Child Maltreatment report by the Department of Health and Human Services, 17% of maltreated children were victims of physical abuse, though it is suspected that the true value is higher due to suspected underreporting or under recognition. Younger children are at highest risk of abuse. The child fatality rate in the United States is estimated at 1.05 children per 100,000 again with the youngest children being the most often affected. Clinicians who provide care for children should be competent in eliciting a detailed history when an injured child presents for care and there is a concern for physical abuse. The history will be the basis of determining whether a fracture is consistent with the biomechanics of the morphology and developmental capabilities of the child. Clinicians should also be knowledgeable about patterns of skeletal injury suggestive of physical abuse to ensure that the recommended medical evaluation is timely, and notification of child protection agencies is promptly initiated for the determination of a safe disposition when necessary.
History and physical should also include screening questions relevant to diseases or conditions that may cause a child to have a predisposition to fractures. Whether the fracture is related to inflicted injury versus neglectful supervision should also be considered. Indicators of neglect may be elicited during the psychosocial assessment.
The presence of racial disparities in the evaluation, reporting, and case substantiation rates of abuse among certain minority children is a topic that has been repeatedly confirmed in the literature. In fiscal year (FY) the incidence 2022, American Indian or Alaska Native children have the highest rate of victimization at 14.3 per 1,000 children in the population and Black or African-American children have the second highest rate at 12.1 per 1,000 children. Lane and colleagues investigated the racial difference in the evaluation of pediatric fractures and found that minority (Black and Hispanic) children in the study’s cohort had a higher rate of abuse diagnosis. When the researchers controlled for the likelihood of abusive injuries in minorities, they were still found to be more likely to be evaluated and reported for suspected abuse. The study’s conclusion included an assessment that “racial differences do exist in the evaluation and reporting of pediatric fractures for child abuse, particularly in toddlers with accidental injuries.” In another study, results showed that minority children and children without private insurance were evaluated with skeletal survey (SS) more often than white children and children with private insurance despite lack of evidence to support this practice.”
Current evidence
Manifestation of Abusive Fractures
In children who have been physically abused, presentation with a fracture is the second most common physical finding, following cutaneous injury. Abusive fractures occur in an estimated 11% to 55% of children who are physically abused.
The youngest children have the highest rate of abusive fractures. When young children with limited or no independent mobility present with fractures, physical abuse should be strongly considered. , In one study, approximately 80% of all abusive fractures occurred in children aged less than 18 months. In a similar study, Leventhal and colleagues studied hospitalized children aged less than 36 months and found that 24.9% of fractures attributable to abuse occurred in children aged less than 1 year (incidence of 36.1 per 100,000). Incidence of abusive fractures decreases with age and achievement of mobility milestones. In that same study, rate of abusive fracture in children aged 12 to 23 months decreased (7.2%) and continued to decline in children aged 24 to 35 months (2.9%).
The presence of multiple fractures is also associated with an increased likelihood of abuse , , and approximately one-third need orthopedic treatment. Worlock and colleagues found that 74% of abused children had 2 or more fractures compared with 16% of non-abused children. No child in the accidental cohort of this study had more than 2 fractures. If associated significant bruising was present, it most often involved the head and neck.
Abusive fractures in young children most commonly affect the longs bones, skull, and ribs. , , Table 1 outlines the proportion of abusive fractures in each bone and categorizes them by age.
| 0–11 mo | 12–23 mo | 24–35 mo | |
|---|---|---|---|
| Skull | 17.1 | 8.6 | 27.6 |
| Ribs | 69.4 | 28.5 | 4.7 |
| Clavicle | 28.1 | 16.7 | 4.7 |
| Humerus | 43.1 | 6.8 | 1.6 |
| Femur | 30.5 | 4.8 | 2.5 |
| Radius/ulna | 62.1 | 19.8 | 6 |
| Tibia/fibula | 58 | 16.1 | 3.7 |
Fracture Patterns and Specificity for Abuse
Though multiple studies have shown that some fracture patterns are highly correlated with physical abuse, , any fracture can be the result of abuse and none are pathognomonic. Conversely, fractures that commonly result from accidents can also be caused by abuse. , Diagnosis of abuse should not be based solely on an individual fracture but take into account the child’s full evaluation. The specificity of fractures for child abuse has been outlined and published Kleinman’s textbook “Diagnostic imaging of child abuse” and is commonly referenced in the literature and used in clinical practice as one of the cornerstones for abusive fracture assessment.
Biomechanics
In children, the bones that are commonly fractured and which fracture morphologies are produced are different from what is seen in adults. The structural and mechanical properties of the developing skeleton allow for a higher degree of pliability that causes a different response to trauma than what is seen in a mature skeleton.
Fracture morphology reflects the force applied to the bone, the bones response to that force, and the ability of the bone to withstand the force applied. A detailed explanation of the injury will allow the clinician to make a more informed assessment about the likelihood that the reported mechanism caused the fracture morphology or fracture pattern observed. The biomechanics, example mechanisms, and radiographic appearance of abusive fractures as described by Pierce and colleagues and Kleinman are summarized in Table 2 .
| Fracture Morphology | Biomechanic Description | Radiographic Description | Example Injury Mechanism | Key Points |
|---|---|---|---|---|
| Transverse | Failure under co-occurring tensile and compressive modes; direct force causing bending, indirect force causing tension loading | Fracture line perpendicular to the long axis of the bone | Direct blow to an extremity with an object Bone impacted directly against a corner, for example, co-fall with an extremity between parent and table edge Indirect tension with ankle inversion | Often related to high energy mechanisms Comminution with progressively increasing forces |
| Greenstick | Bending perpendicular to the long axis, gradual change in the contour | Curved distortion/diaphyseal angulation without a discrete fracture line | Same mechanism as transverse fractures but with force below fracture threshold | |
| Spiral | Twisting about the long axis of the bone | “Corkscrew” | Abusive “girip and twist” injury (Worlock, Stower and Barbor 1986) accidental “plant and twist” slip and fall while running | Highly correlated with abuse in infancy Not pathognomonic for abuse,in ambulatory children, consider “toddler fracture”; Forces can be indirectly distributed to the femur Uncommon in falls >10 ft |
| Oblique | Combination of loading forces applied at a 30-45 degree angle to the long axis of the bone Torsion compression bending Morphology reflects the primary load type | Long oblique TORSION +compression; fracture line involves more of the diaphysis Short oblique Torsion+ COMPRESSION ; shorter fracture line “Butterfly fragment” compression while bending+torsion | A fall on a bent knee with a twist as the knee hits the floor | Distinguishing spiral from oblique may change based on radiographic angle; not distinguishable at times |
| Buckle “Torus” | Axial load along the long axis of the bone at the junction of the metaphysis and diaphysis When the mid-diaphysis is the site of fracture, bending rather than buckling likely occurred; unicortical/incomplete fracturing may result | Bulging of the cortex | Falling onto a bent knee or elbow Infant being thrown or slammed down onto a hard surface | Transition zone between metaphysis and diaphysis susceptible to injury; this morphology is unique to developing bone Usually at the proximal or distal 1/3 of the bone at metaphysis “Torus” implies circumferential cortical buckling |
| Compression (spine) | Axial load along the long axis of the spine | Vertebral loss of height, Wedged appearance of the vertebrae | Forceful slamming into a seated position violent shaking causing hyperflexion or hyperextension of the spine | Thoracic, lumbar, sacral spine |
| Classic Metaphyseal Lesion | Torsion, traction, shear strains applied across the metaphysis | Appearance is based on the radiologic projection through the bone and extent of bone failure; corner, chip, bucket handle | Forceful yanking or pulling of an extremity; flailing of a limb during violent shaking; slamming a child on a surface while being held by an extremity | Higly correlated with abuse in infancy Healing occurs without periosteal new bone formation (Kleinman, Marks, and Blackbourne 1986) |
| Epiphyseal separation | Similar mechanism as CML | Displacement of the cartilaginous portion of the bone | Similar mechanism as CML | Magnitude of force determines if a CML or epiphyseal separation is caused; most severe force cause epiphyseal separation Most occur with acciddents In abused infants most common in the distal humerus, also involves proximal humerus and femur Consider birth trauma in newborns US and MRI aide in diagnosis due to large cartilagenous component |
| Subperiosteal new bone formation | Shear and torsional load applied along the outer bone surface parallel to the long axis of the bone causing periosteal separation and hemorrhage | Thin or hazy cortical bone separated from the original cortex by a thin dark line (hemorrhage) | Shaking, grabbing or yanking an extremity, slamming the infant onto a surface | Not a discrete fracture but may represent healing bone injury Most commonly involves the femur and humerus non specific and may represent a physiologic process; suggested when present bilaterally and restricted to the diaphysis |
High Specificity Abusive Fractures
Classic metaphyseal lesions
The classic metaphyseal lesion has long been considered the injury with one of the highest specificities for abuse and is the long bone injury most often encountered in association with infant mortality from physical abuse. In a postmortem study of children who died from physical abuse, 89% of the long bone fractures involved the metaphysis. Walker and colleagues did an updated literature review to assess whether the literature continued to support the high association of classic metaphyseal lesions (CMLs) and rib fractures with physical abuse. Results showed that these fracture types continued to have a significant correlation with physical abuse.
The distal metaphysis is vulnerable to injury due to its lack of chondrocytes, immature cell organization, and less calcification making it “softer” in comparison to the adjacent diaphysis and the attached tendons and ligaments.
CMLs can occur in any long bone but are most commonly observed in the tibia, distal femur, and proximal humerus and are rarely found in the proximal femur. , ,
Rib fractures
Rib fractures in infants are also highly associated with physical abuse comprising 5% to 27% of all abusive fractures in one study. Due to known limitations in identifying acute rib fractures radiologically, the true incidence of rib fractures is thought to be even higher. Rib fractures can be difficult to detect on initial chest radiographs and are commonly occult findings due to presence of overlying bones or soft tissue and if there is lack of fracture displacement. Rib fractures are more likely to become apparent on follow-up radiographic imaging after healing with callus formation can be appreciated. Kleinman and colleagues found a higher incidence of abusive rib fractures (51%), using postmortem radiologic and histopathological studies.
The association of rib fractures with abuse is related to the unique anatomic features of infants and the type of forces needed to cause this fracture type.
An infant’s rib cage has a higher pliability than adults due to the increased proportion of cartilage; thus, it can tolerate higher degrees of deformation before a fracture occurs. Due to the increased amount of force needed to cause a rib fracture, when present, rib fractures suggest that the child has experienced a significant amount of force applied to their chest.
In one meta-analysis, there was a 77% incidence of abuse in children aged less than 3 years who presented with rib fractures. In a similar meta-analysis, the probability for abuse was 71% for rib fractures, which, in that study, was the highest probability for abuse of all fracture types. In the same study, children with rib fractures from suspected abuse had more total rib fractures than children with rib fractures from other causes like motor vehicle collision (MVC), documented violent chest trauma, post-surgery, or other medical explanations.
In a systematic review by Paine and colleagues , rib fractures in children aged less than 1 year were the only characteristic of rib fractures significantly associated with an increased likelihood of abuse (91% prevalence; 66%–100% prevalence for children <2 y) when MVC and bone pathology were excluded.
Rib fracture distribution
Abusive fractures can occur at any point along the rib arc though multiple studies demonstrate a tendency for abusive rib fractures to involve the posterior and lateral ribs. , Abusive rib fractures have also been shown to involve adjacent ribs—most often in mid-level locations, and they are more commonly left sided. , In 2019 meta-analysis, rib fracture location was not shown to be associated with likelihood of abuse. In a study by Raynor, 56% of abusive fractures were rib fractures, the median number of fractures per infant was 6, fractures in at least 2 adjacent ribs occurred in 19 out of 21 infants, and more fractures occurred on the left side with more frequent involvement of ribs 4 to 7 in the lateral and posterior-lateral segments. The seventh rib was most commonly fractured rib. Kriss and colleagues found that 81% of abused children had rib fractures, 67% were left sided, and the fractures were most often in the posterior and lateral regions. Ribs 5 to 8 were frequently involved and 54% of the children with rib fractures had other skeletal fractures.
Common in Child Abuse, but Low Specificity for Abuse
Long bone injuries
Long bone fractures are the most common fracture type in abused children outside of infancy. After 2 years old, long bone fractures are more likely to be the result of accidental trauma due to a child’s increased ability to walk, run, and cause injury to themselves.
Femur fractures
The femur is the most commonly fractured long bone from abuse after infancy. Shaft fractures are most common, followed by fractures of the distal third. Proximal femur fractures are uncommon in accidental and non-accidental trauma. Without a history of high-velocity trauma, this location should raise suspicion for non-accidental trauma (NAT). Spiral femur fracture is the most common morphology associated with abuse in children aged less than 15 months. No significant difference is seen in children aged older than 15 months. Femur fractures are typically associated with soft tissue swelling and limitation in extremity movement.
In a systematic review by Wood and colleagues, rate of abuse was 16% to 35% in children aged less than 12 months but dropped to 1.5% to 6% in children aged greater than 12 months. In this study, the fracture location and type were not significantly associated with abuse with the exception of an increase in abuse likelihood with distal metaphyseal fractures. This study highlights the correlation among the presence of additional injuries, nonambulatory status, age less than 12 months, and provision of a suspicious history (no history of trauma, unwitnessed trauma, or inconsistent mechanism reported) as being significantly correlated with the likelihood of abuse.
Tibia fractures
The tibia is the third most frequently injured long bone in abused children following the femur and humerus. Tibial fractures are worrisome for abuse in non-weight-bearing children without a history of significant trauma.
CMLs are more common in the tibia than in any other site. The distal and proximal portions are often involved in comparison to the femur, where CMLs are uncommon proximally. The most common type of tibia fracture changes based on the age group. In older children, tibia fractures are related to high-energy trauma, while in younger children, low-energy falls or twisting of the leg about the long axis of the bone can result in an accidental “toddler fracture.” A history of a “slip and fall” in an ambulatory child who recently began walking is a history commonly provided explanation.
In a study about the prevalence of abuse in trauma patients aged less than 18 years with lower extremity fractures, 41 of 55 lower extremity fractures were linked to abuse. Femur fractures were most common, followed by tibia fractures.
Humerus fractures
Humerus fractures are highly correlated with abusive in young children due to the arms being a common site of forceful manipulation during assault. The most common location of abusive fractures of the humerus is the middle and distal thirds of the shaft. Most common morphologies are spiral or oblique. In one study, the overall estimate of abuse probability in cases of suspected and confirmed abuse in children aged less than 3 years was 54%.
Presentation with soft tissue swelling and decreased arm movement are common though bruising is often absent as is the case with many abusive fractures. The humerus is a likely bone for incidental finding of CML in infants not initially suspected of abuse who undergo chest imaging for other reasons. This is also the case in the incidental identification of rib fractures.
Radius and ulna
Adjacent transverse fractures of the shafts of the radius and ulna have been called “nightstick” injuries and are caused by a blow to the forearm. In older children, this fracture pattern can be the result of a child attempting to block a blow to the face or upper body. Indirect forces from bending can also cause this presentation. In the distal forearm, buckle fractures and transverse fractures of the metaphysis are more difficult to distinguish from accidental injury.
Skull fractures
Simple linear parietal skull fractures are the fracture type most common in accidental and non-accidental trauma. , , In one study, the mean age at presentation was 5 months. The most common history provided was a fall and the second most common history was unknown mechanism. Overall, 62% of patients had no intracranial hemorrhage (ICH). In another study, a history of fall was provided for 70% of cases; greater than 80% of cases involved the parietal bones. Bilateral simple linear fractures occurred in 79% of cases of accidents versus 35% in abuse. Among the cases of abuse, complex fractures were present in 55% of cases versus 21% of accidental cases. Features that were more common in abusive cases were displaced fractures, diffuse intracranial hemorrhage, skin trauma, and additional fractures on SS.
In one retrospective review whose aim was to build on what is known about abusive head trauma, simple linear parietal skull fractures were the most common. Complex skull fractures, when present, were more highly associated with abusive head trauma.
Clavicular fractures
The clavicle is the most commonly fractured bone in children with midshaft fractures occurring in up to 80% of cases. This fracture type can be associated with accidents, birth trauma, and abuse.
Moderate Specificity
Though rare, vertebral body, hand, and foot fractures may be the only indication of abuse or may provide additional information that supports the diagnosis of abuse in the setting of other identified injuries.
Evaluation
Diagnostic Imaging
Skeletal survey
Both the American Academy of Pediatrics (AAP) and the Royal College of Paediatrics and Child Health (RCPCH) recommend evaluation by SS for children aged more than 2 years in whom physical abuse is suspected as a means of evaluation for clinically unapparent fractures. Components of a SS as recommended by the AAP are outlined in Table 3 .
