Physical child abuse: Diagnostic evaluation and management

INTRODUCTION — The diagnostic evaluation for physical child abuse will be reviewed here. Recognition of physical child abuse, the orthopedic aspects of child abuse, and the evaluation of abusive head trauma are discussed separately:

●(See "Physical child abuse: Recognition".)

●(See "Orthopedic aspects of child abuse".)

●(See "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children".)

RATIONALE — The identification of suspected abuse at the initial evaluation is critical to treat the injuries and to protect the child from a subsequent, perhaps more serious injury [1-3]. Children returned to their families/primary caregivers without intervention after an event of maltreatment have an 11 to nearly 50 percent chance of a second event [4-9]. The majority of both initial and recurrent events are neglect. In a study of children seen in an emergency department for accidents that were retrospectively determined to be physical abuse, 13 percent of patients suffered another physical abuse event in the subsequent five years [10].

Despite the urgency of identifying suspected child abuse and the existence of mandatory reporting laws throughout the United States and in many countries around the world, child maltreatment remains consistently under-identified and underreported [11,12]. Multiple factors contribute to this problem, including the knowledge and expertise of health care practitioners as well as variation in interpretation of what constitutes a reasonable suspicion for reporting [13-15].

INITIAL STABILIZATION — Traumatic injuries warrant assessment and management based upon the patient's level of stability and according to the principles of Advanced Trauma Life Support. (See "Trauma management: Approach to the unstable child", section on 'Initial approach' and "Approach to the initially stable child with blunt or penetrating injury".)

APPROACH — It is rarely possible (and almost never necessary) to definitively conclude that physical abuse has occurred in the first few hours of an evaluation. When physical child abuse is suspected, clinicians should emphasize the need for further evaluation and avoid making accusations. Examples include: "Sometimes, when we see injuries like this, there are other injuries that can put a child at risk, and we need to test for them," or "This is more injury than we would normally expect from the event you are describing. We should be careful to determine whether there is something else medical going on, and to make sure that no one is hurting the child."

When questions of abuse arise, a thorough evaluation must assess the plausibility of the given history (table 1), findings on physical examination (table 2), the possibility of undiagnosed explanatory medical conditions, and the presence of additional occult injuries that further contribute to the diagnosis of abuse. (See "Physical child abuse: Recognition".)

Red flag injuries that raise suspicion for physical child abuse and warrant additional radiologic and laboratory evaluation vary according to the age of the child and the specific injury, as shown in the table (table 2).

When evaluating the child who is suspected of being physically abused, it is also important to remember that findings that appear to indicate abuse may result from other causes. Familiarity with the medical conditions and cultural practices that mimic child abuse and careful medical, laboratory, and radiologic evaluation can facilitate arrival at the correct diagnosis, initiation of appropriate therapy, and avoidance of the consequences of an unfounded report of suspected child abuse. (See "Differential diagnosis of suspected child physical abuse", section on 'Cultural practices'.)

Ancillary studies recommended for the evaluation of physical child abuse, as elaborated below and in American Academy of Pediatrics policy statements [16], are complex with multiple contingencies, primarily the child's age and type of injury (table 3). As a result, compliance with best practices is inconsistent [17-19].

Use of a paper or electronic checklist has resulted in improved screening and increased reporting of suspected child abuse [17,20-23]. These interventions have directed actions from full examination and involvement of social services to complete order sets of occult injury and medical condition screening. For example, a standardized electronic order set for nonaccidental trauma based upon a best practice institutional guideline increased adherence to recommended testing from 47 to 69 percent [20]. In one trial, use of a physical abuse order set was associated with full compliance with institutional clinical guidelines for testing, although the automated alert system for activation of the order set did not increase overall compliance with recommended testing [21]. However, compliance was 84 percent at baseline [21]. As a result, while acceptance and use of the checklist was high in this study, it was not possible to document increased compliance with recommendations. Additional measures, such as a periodic multidisciplinary conference to assure understanding of and encourage compliance with standardized practices may yield additional benefits [24].

EVALUATION — The evaluation of suspected child abuse begins with obtaining a complete history, performing a physical examination, and observing the interactions between the parents/caregivers and the child.

History — Once an injury is identified, the history should identify an explanatory trauma or an explicit denial of trauma. A detailed history of trauma events should include preceding activity elevation and motion, events leading to the trauma event, mechanics of the injurious circumstances, and subsequent actions and symptoms of the patient. This sort of detail is necessary for evaluating the plausibility of the reported history [25]. While an accurate summary or paraphrasing of given answers may be adequate, when particularly important statements are made, quoting the exact words of the reporter will avoid concerns of inaccurate interpretation by the clinician and serve any subsequent legal proceeding better.

An acute injury presents the ideal reason for asking a child with adequate language skills to explain "what happened to you." Ideally, this would be done away from adults whose presence might influence the child. In the setting of acute trauma, separating the child might not be practical for a multitude of reasons. The choice of deferring child questioning or asking in the presence of others must be individualized as is further discussed below. Spontaneous comments by the child must, however, be accurately recorded along with the situation and statements that provoked them.

The desire to interview a child directly is often weighed against the risk that age-inappropriate questioning can yield misleading information. Children will attempt to answer questions they do not understand, and repeated questioning can be perceived to imply that the first answer was not correct. Coordination between medical providers to avoid repeated questioning and accurate "sign out" of earlier child statements may decrease repeated history-taking.

A well-prepared physician taking a medically oriented history is likely to obtain information not obtained in forensically oriented interviews, particularly in older children. Awareness of forensic interviewing principles and techniques is important preparation before clinicians embark on this task. Because a child's initial statements have critical importance to the medical evaluation, and because some children are never questioned by any investigator, clinicians should consider talking to children with well-developed linguistic capacity [26]. Because obtaining a clinical history from a child may create inconsistencies with forensic interviews, particularly in younger children, coordination between medical care systems, children's protective services, and local law enforcement, as exemplified by the Child Advocacy Center model, should be planned. When a child's history is taken, to the degree possible, it is best to report both the questions asked and the answers provided by the child as literally as possible and, ideally, in exact quotes.

Regardless of who obtains the history, the process should begin with open-ended questions about how the injuries were sustained and then proceed to more specific questions [27]. The questions should be directed first to the child if they can speak. The author usually interviews the parents/caregivers and child together initially; if a significant suspicion of abuse is raised, permission is sought from the parents/caregivers to interview the child alone. Most parents/caregivers agree to this; whether or not it is permitted is documented in the medical record. Refusal to permit the child to be interviewed alone is considered a "red flag" for abuse. Social services personnel are helpful in this situation, both in assessing family dynamics and in facilitating the removal of the parents/caregivers from the examination room (to interview them separately).

Extensive research on interviewing children stresses the importance of using open-ended questions that do not introduce concepts of abusive acts or an abuser [28]. Requests for unstructured narrative are best (eg, "Tell me about your boo-boo") followed by "WH" questions. For example, following a child's statement that "they hurt me":

●"You said 'they' hurt you, who is 'they'?"

●"What did they do?"

●"Where did they do it?"

●"When did this happen?"

Interviews should consider a child's age and developmental ability. While children as young as three years old can often supply answers to "who" and "what" questions, "where" and especially "when" comprehension comes later. If a young child has difficulty answering "when" questions, other details can suggest timing. For example, a child who cannot tell you that the abuse occurred two weeks ago may be able to report that the abuse occurred when they visited a neighbor's house, or when Christmas decorations were still up, or when a particular show was on television.

Clinicians should avoid asking the child to select from lists or answer yes/no questions because these approaches are inaccurate [28].

Historical factors that raise the suspicion of abuse include (table 1) (see "Physical child abuse: Recognition", section on 'Red flag history'):

●The history provided by the parent or caretaker is inconsistent with the injuries of the child (eg, a child with a history of falling from a sofa onto a carpeted floor who presents with multiple fractures, liver laceration, or traumatic brain injury).

●The history is vague or lacking in detail.

●The history changes in repeated versions given by the same caretaker to different health care workers, or conflicting histories are given by different household members (eg, infant was burned while bathing in the sink/infant was burned while bathing in the bathtub; mother was bathing infant/sister was bathing infant).

●The injury is attributed to actions of young siblings; this account may be a cover story or may be true and related to sibling rivalry, inadequate supervision, or violence in the home. (See "Child neglect: Evaluation and management" and "Intimate partner violence: Childhood exposure".)

●No history is offered (eg, "I don't know what happened; their leg was just suddenly broken").

●The history is inconsistent with the developmental stage of the child (eg, a four-month-old infant who turns on the hot water in the bathtub and is burned).

●The history is implausible (eg, a child presents with swelling of the thigh, with a history given by the parents/caregivers that "a roach bit them"; radiographs reveal a spiral fracture of the femur).

●There is a history of prior bruising or orofacial injury in an infant who is not cruising.

If the patient is not known to the examining clinician, information about past injuries, hospitalizations, and emergency department visits should be obtained, as well as information regarding the child's regular source of primary care and past medical history, including illnesses or conditions that mimic child abuse. (See "Differential diagnosis of suspected child physical abuse".)

It is often helpful to contact the child's primary care provider to ascertain whether or not the child has received routine health care and immunizations as scheduled and whether there have been prior concerns of child abuse.

Some authors have stressed the importance of the social history and child indicators of past distress to increase sensitivity to child abuse [29,30]. Such an approach must be tempered against the possibility of increasing social biases. In a study of consensus on best practices, child abuse pediatricians considered a history of child care settings to be a required element, and past history of abuse to the index child, intimate partner violence, past social services involvement, and abuse experience of the caregiver during childhood as "highly recommended" [31].

Physical examination — Evaluation of the general appearance of the child should include assessment of their clothing (eg, is it appropriate for the season? is it clean and in good repair?). The physical examination ideally is performed with all of the child's clothing removed; the clothing can be removed sequentially (eg, examination of the upper extremities and torso while the lower body is clothed, followed by examination of the perineum and lower extremities with the torso clothed).

Characteristic skin lesions, swelling or deformity, bone tenderness, or reluctance to use an extremity should be sought specifically, as should retinal hemorrhages, trauma to the genitals or mouth (eg, lingular or labial frenulum tears or palatal petechiae in a pre-mobile infant), and signs of neglect (eg, malnourishment, poor hygiene). (See "Physical child abuse: Recognition", section on 'Red flag physical findings' and "Child abuse: Eye findings in children with abusive head trauma (AHT)" and "Child neglect: Evaluation and management".)

Findings that raise the suspicion of child abuse include (table 2) (see "Physical child abuse: Recognition", section on 'Red flag physical findings'):

●Injuries with patterns that indicate a method of infliction (figure 1 and figure 2):

•Slap, belt, loop of cord, and other shaped bruises

•Cigarette, iron, spatula, and other shaped burns

•Immersion burns up to a "high tide" level or accompanied by symmetrical sparing of the buttocks where they contacted the bottom of the tub, excluding contact with the heated water

●Multiple fractures in various stages of healing or different types of injuries coexisting (eg, bruises, burns, and fractures)

●Injuries with a high epidemiological association with abuse [32,33]

•Bruises of the trunk, ear, neck, angle of the jaw, fleshy cheek, and eyelid (see "Physical child abuse: Recognition", section on 'Inflicted bruises')

•Bruises in children who cannot cruise (see "Physical child abuse: Recognition", section on 'Inflicted bruises')

•Frenulum tears and subconjunctival hemorrhages, especially in children <2 years old and those who are not yet walking independently

•Long-bone fractures in children who do not walk (see "Orthopedic aspects of child abuse", section on 'Femur fractures')

•Rib fractures in infants younger than one year of age (see "Orthopedic aspects of child abuse", section on 'Rib fractures')

•Subdural hematoma in infants younger than one year of age (see "Physical child abuse: Recognition", section on 'Abusive head trauma')

●Hollow viscus injury in children younger than four years of age (see "Physical child abuse: Recognition", section on 'Visceral injuries')

●Injuries that are epidemiologically or biomechanically unlikely to arise from the reported trauma event

●Evidence of poor caretaking (a child who is dirty or inadequately clothed) may raise suspicion of abuse; however, these factors correlate more strongly with neglect or poverty than with abuse, and abuse may be present in the absence of these signs

●Sudden onset of altered mental status not attributable to medical illness (eg, hypoglycemia, hypoxemia, shock) or other signs of poisoning (table 4)

●Injury to the genitalia

While a careful physical examination can raise the level of concern for abuse, many abusive injuries remain hidden, even from a careful examination. Occult abdominal injuries [34-37], fractures [38], and brain injury [39-41] have been well described in children with normal examinations.

Family interaction — The behavior of the parents/caregivers and the interaction between family/household members should be observed carefully during the evaluation of the child. Certain behaviors and/or types of interaction may increase the level of suspicion for child abuse. Such behaviors include:

●Arguing, roughness, or violence

●Aloofness and lack of emotional interaction between parents/caregivers or between parents/caregivers and children

●Negative characterization of the child by the parent [42]

●Inappropriate response to the severity of the injury (eg, lack of appropriate concern)

●Inappropriate delay in seeking medical care (see "Physical child abuse: Recognition", section on 'Delay in seeking care')

●A partial confession by the parent (eg, "I hit them, but not that hard") or a frank admission by parent that injury was inflicted; such confessions occur occasionally and are an indication that the parent realizes that abuse is a problem and is seeking help

Laboratory studies

Trauma assessment — Children with serious traumatic injuries should undergo emergency laboratory studies and imaging appropriate for the evaluation of multiple traumas as discussed in detail separately. (See "Trauma management: Approach to the unstable child", section on 'Laboratory studies' and "Trauma management: Approach to the unstable child", section on 'Screening radiographs' and "Trauma management: Approach to the unstable child", section on 'Adjuncts to the secondary survey'.)

Initial physical abuse evaluation — The laboratory evaluation for child abuse should be tailored to the findings on physical examination as well as to the specific nature or history of the injury (table 3) [16].

Evaluation for bleeding disorders — An approach to deciding whether to obtain an evaluation for bleeding disorders in suspected child abuse is provided in the algorithm (algorithm 1) [43]. While some child abuse pediatricians will recommend these studies in all cases, it may be reasonable to omit them when the diagnosis of abuse is more secure due to witnessed abuse, confessed abuse, skin injuries with the imprint of an object or hand, or multiple injuries not explainable by a single medical condition.

An evaluation of the child's coagulation may be warranted either to detect the coagulopathy that often results from traumatic brain injury or to test for coagulopathy that could be the underlying cause of bruising, retinal hemorrhage, or intracranial bleeding, rather than child abuse. At a minimum, bleeding studies should include a complete blood count with differential and platelet count, and prothrombin and partial thromboplastin times. A more comprehensive approach might include fibrinogen, thrombin time, Factors 8, 9, 11, and 13, and von Willebrand testing, though many specialists choose to consult hematology before ordering more advanced testing [16,44]. (See "Approach to the child with bleeding symptoms".)

Anemia with an abnormal peripheral smear and cell indices may be one manifestation of neglect-associated nutritional anemia. Alterations of prothrombin and/or partial thromboplastin time are present in bleeding disorders (eg, factor deficiencies, significant hepatic injury, or warfarin or anticoagulant rodenticide poisoning) or in disseminated intravascular coagulation (DIC) resulting from overwhelming injury or sepsis. In the setting of DIC, bleeding time, fibrinogen level, and thrombin time may be useful as well. (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults".)

Screening studies — Additional laboratory studies to identify occult injuries or medical conditions should be obtained based upon patient characteristics (table 3):

●Serum ALT and AST – When elevated (>80 U/L) after trauma, serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) suggest injury to the liver. The clinician should have a low threshold for obtaining these studies in young children who cannot reliably communicate the presence of abdominal pain or tenderness and those with other significant injuries (such as intracranial injury or rib/long-bone fracture). Patients with elevated ALT or AST warrant appropriate abdominal imaging for intraabdominal injury. (See 'Abdominal injury' below.)

In children who undergo evaluation for physical abuse, elevation of either the AST or the ALT may indicate intraabdominal injury, even in patients with no clinical signs of abdominal trauma [16]. For example, in a multicenter observational study, 17 of 54 children with identified intraabdominal injury had no clinical findings. Fourteen of 17 had elevations in liver enzymes [34]. In a follow-up study, this threshold had a sensitivity of 84 percent and a specificity of 83 percent for occult intraabdominal injury [37].

●Serum lipase – When present, an elevated serum lipase supports traumatic pancreatitis. However, a normal serum lipase does not exclude intraabdominal injury. Serum amylase is less specific for pancreatic pathology.

●Testing for metabolic bone disease – In children with fractures as the sole manifestation of possible physical abuse, further evaluation may help exclude bone disease (eg, osteogenesis imperfecta, rickets, hypophosphaturia) as the underlying etiology of fractures. While some child abuse specialists will recommend these studies in all cases, it may be reasonable to omit them when the diagnosis of abuse is more secure due to witnessed abuse, confessed abuse, skin injuries with the imprint of an object or hand, or multiple injuries not explainable by a single medical condition. In cases where fractures are the only cause for concern for abuse, it is reasonable to consult with an expert in bone metabolism, such as an endocrinologist or geneticist, and to screen for disease by obtaining [16,45,46]:

•Serum calcium or ionized plasma calcium, serum phosphorus, and serum alkaline phosphatase

•Intact parathyroid hormone level

•25-hydroxy vitamin D level

•Osteogenesis imperfecta genotype or phenotype testing (see "Osteogenesis imperfecta: An overview")

If there are concerns for copper deficiency or Menkes disease based upon physical findings, measuring levels of copper or ceruloplasmin may be warranted, especially in male infants younger than six months of age.

It is possible for testing of rare differential diagnoses to multiply the number of ordered tests dramatically. In the absence of clinical indicators for these conditions, the result is likely to be more false positives or variants of unknown significance than a true positive. Having a standardized approach can prevent this problem, improve consistency of evaluations, and promote efficiency. For example, one regional referral institution developed a standardized approach for metabolic and genetic consults in suspected child abuse [47]. For fractures, genetic testing of COL1A1, COL1A2, IFITM5, and in children of West African descent P3H1 was recommended. For head injury plasma acylcarnitine, free/total carnitine, urine organic acids, and blood 3-OH glutaric acid, along with genetic testing of GCDH (associated with glutaric acidemia, type I) was recommended.

Many of these tests are not rapidly available. Mandated reporting of suspicion for child abuse should not be delayed while awaiting consultation or results of metabolic testing in infants and children with fractures that have high specificity for abuse (eg, rib fracture, metaphyseal corner fractures, multiple fractures in various stages of healing). (See "Orthopedic aspects of child abuse", section on 'Overview'.)

Further testing for bone fragility is an area of some debate. Several theories of transient bone fragility have been proposed in the academic literature and in legal settings. While some clinical entities such as rickets and osteogenesis imperfecta are well recognized [48-50], other proposed entities such as Ehlers Danlos syndrome and temporary brittle bone disease have been widely questioned or discredited [51-53].

●Other studies – Serum and urine electrolytes and osmolality should be measured in the settings of abusive head trauma, dehydration caused by sustained forced exercise, or water intoxication caused by forcing of fluid ingestion or by drowning [54-56]. (See "Hypernatremia in children" and "Hyponatremia in children: Etiology and clinical manifestations".)

A urinalysis may be useful for children with clinical findings that raise concern for renal injury. However, absent clinical features of abdominal or renal injury, urinalysis is not a valuable screening test for occult abdominal injury and, in those patients, should not be obtained solely for that purpose. For example, in a series of 237 children undergoing evaluation for suspected child abuse, urinalysis had 0 percent sensitivity for occult abdominal injury; all children with positive hematuria screen had either clinical indications of abdominal injury, or elevated AST/ALT [57]. (See "Pediatric blunt abdominal trauma: Initial evaluation and stabilization".)

Rhabdomyolysis can be caused by extensive muscle injury, and evaluation is warranted in patients who have extensive bruising or urine that is dipstick positive for heme but lacks red blood cells on microscopic examination (algorithm 2 and table 5) [58,59]. Patients with rhabdomyolysis typically have pigmented granular casts in the urine, a red to brown color of the urine supernatant, and a marked elevation in the plasma level of creatine kinase [60,61]. (See "Rhabdomyolysis: Clinical manifestations and diagnosis" and "Urinalysis in the diagnosis of kidney disease", section on 'Red to brown urine'.)

Rhabdomyolysis may lead to oliguria, acute tubular necrosis, and renal failure [58,59]. Blood urea nitrogen and creatinine usually are elevated. Other laboratory abnormalities associated with this condition include hypocalcemia, hyperphosphatemia, and hyperuricemia. Ten to 40 percent of victims develop hyperkalemia, resulting either from potassium release from damaged cells or from renal compromise [58]. Urographic contrast agents should be avoided because they may precipitate or worsen renal failure. The management of rhabdomyolysis involves aggressive hydration and diuresis [58] and is discussed in detail separately. (See "Rhabdomyolysis: Clinical manifestations and diagnosis" and "Prevention and treatment of heme pigment-induced acute kidney injury (including rhabdomyolysis)".)

Toxicology — Toxicology screens (serum and/or urine drug panels) should be obtained in children with suspected poisoning or drug exposure and in those who have symptoms compatible with drug toxicity (table 4), brief resolved unexplained event (BRUE), or unexplained neurologic symptoms (eg, ataxia) [27,62-65]. In addition, toxicologic analysis beyond the typical rapid drugs of abuse screens are often needed to establish the diagnosis of malicious poisoning. As an example, in one series of 274 infants with apparent life-threatening events, comprehensive toxicology screens identified unexplained toxins in nearly 20 percent of children, and 8 percent of these tests identified an agent that could cause apnea or other potentially life-threatening events [63]. It may be necessary to specifically request an assay for the poison or drug of interest when that substance may not be included or detected on the routine blood and urine toxicologic screen performed by the laboratory (eg, antihistamines or oral hypoglycemic agents). In these patients, consultation with a regional poison control center or medical toxicologist is advised to assist with recommendations for specific testing and identification of specialized laboratories where such testing can be performed. (See "Medical child abuse (Munchausen syndrome by proxy)" and 'Additional resources' below.)

If a screening test (eg, enzyme-mediated immunoassay for drugs of abuse) is positive, then a more definitive test should be performed to confirm and quantify the amount of substance present (eg, gas chromatography/mass spectrophotometry). (See "Testing for drugs of abuse (DOAs)".)

Specialized toxicology screening is also indicated during autopsies performed as part of the evaluation of a sudden unexpected infant death. (See "Sudden unexpected infant death including SIDS: Initial management", section on 'Case investigation'.)

The use of screening urine toxicology in potential physical abuse and "drug-endangered children" without signs of intoxication has also been explored with early encouraging results [66,67]. (See "Acute events in infancy including brief resolved unexplained event (BRUE)", section on 'Evaluation of infants not meeting low-risk criteria' and "The substance-exposed child: Clinical features and diagnosis".)

Malicious or deliberate poisoning is a form of child abuse and is defined as the administration of medications or other substances with the intent to harm the child. Commonly implicated agents include opioids, over-the-counter medications, sedatives (eg, benzodiazepines), ethanol, and illicit drugs [68-71]. Infants and children <2 years old are at highest risk. For example, a review of the United States National Poison Data System from 2000 to 2008 found 1439 cases of pharmaceutical child abuse over nine years, including 18 deaths [68]. In this study, "sedating agents" such as over-the-counter antihistamines (eg, diphenhydramine); cough-cold preparations containing antihistamines, dextromethorphan, or codeine; other opioids (eg, methadone or buprenorphine); benzodiazepines; and atypical psychotic agents accounted for half of all exposures; over 90 percent of the fatal exposures involved antihistamines or methadone. In another study of 700 pediatric poisoning-related deaths reported to a United States child fatality database from 2005 to 2018, approximately one-fifth of poisonings were deliberate with over half occurring in infants ≤1 year of age [71]. For all fatal poisonings, a prior history of child maltreatment was present in about one-third of fatalities, and one-sixth had an open Child Protective Services case at the time of death. In addition, opioids and over-the-counter cough/cold preparations were the most commonly implicated substances. A separate report determined that 13 of 40 deaths caused by over-the-counter cough and cold medication poisoning (most commonly diphenhydramine) were caused by intentional administration of supratherapeutic doses designed to sedate or to harm the child [70].

Imaging — Similar to the laboratory evaluation, the radiographic evaluation for suspected child abuse depends upon the age of the child, the presenting complaints, and the physical examination findings (table 3) [72]. When the injuries are unintentional, the radiologic evaluation may provide an alternative explanation (eg, osteopenia, skeletal dysplasia). On the other hand, when the injuries have been inflicted, the radiologic evaluation may identify the extent of injury [72-74]. Guidelines for imaging of children with suspected child abuse have been published by the American College of Radiology [75]. Consensus guidelines have been developed for which injury presentations justify imaging in an "image gently" environment [76-78].

Specific injury patterns associated with child abuse are discussed in detail elsewhere. (See "Physical child abuse: Recognition", section on 'Fractures' and "Orthopedic aspects of child abuse" and "Child abuse: Epidemiology, mechanisms, and types of abusive head trauma in infants and children".)

Skeletal survey — The skeletal survey (table 6) is widely regarded as the best method for detecting fractures in children who have been abused [16,72,79-82]. Orthogonal radiographs (anteroposterior [AP] and lateral) of any areas of bone tenderness, swelling, deformity, or limited range of motion, and all areas that have a history or evidence of previous trauma not otherwise fully imaged by a skeletal survey should also be obtained [72].

●Indications – Concern for physical abuse in children younger than 24 months of age is a strong indication for obtaining a skeletal survey (table 3) [72,79,80]. A consensus panel has also developed a guideline that suggests when a skeletal survey is necessary based upon child age, location, and types of bruising, fractures, and intracranial hemorrhage (table 7) [76-78].

Relative indications for skeletal survey include [72,79,80]:

•Children 24 to 60 months of age with a distracting injury or highly suspicious index fracture (eg, rib or long-bone fracture without history of trauma or metaphyseal corner fractures)

•Concern for abuse in children with impaired mobility or communication skills, or altered level of consciousness at any age

•Children younger than 24 months of age who are asymptomatic but who share a home with an abused child (see 'Identifying injuries in other children at risk' below)

While 24 months has been considered something of a threshold for skeletal surveys in some centers, many guidelines consider the skeletal survey to have utility in older children. As an example, in a multicenter observational study of children referred to a child abuse specialist, 10 percent of children 24 to 36 months of age (33 of 319) had a new fracture identified on skeletal survey [83,84]. In a European study, 32 percent of children with a suspicious index injury had occult fractures detected on skeletal survey: 52 percent with index bruises, 37 percent with head trauma, and 29 percent with a presenting fracture. Rib and tibia-fibula fractures were the most frequently identified [85]. In a United States study of 378 infants aged less than six months with externally evident injury, only 53 percent were appropriately imaged, but 19 percent of those who were imaged had occult fractures [86].

However, a skeletal survey is unlikely to be helpful in verbal, ambulatory children beyond the age of five years. (See "Orthopedic aspects of child abuse", section on 'Skeletal survey'.)

●Technique – A proper skeletal survey requires 21 views (table 6) and special technique as well as interpretation by a radiologist experienced in child abuse imaging. Some families/primary caregivers and clinicians are concerned with the radiation dose of a skeletal survey. While this will differ based on equipment and technique, an estimate of 0.2 mSv for a high-quality survey has been published [87,88]. The risk from this degree of radiation exposure seems significantly lower than the risk of missing evidence of child abuse. If the child will undergo computed tomography (CT) of the head with three-dimensional (3D) reconstruction, then skull radiographs should be omitted from the skeletal survey because head CT with 3D reconstruction has higher accuracy for skull fractures with excellent interrater reliability [89]. (See 'Skull fractures' below.)

A repeat skeletal survey taken two weeks after the initial evaluation may increase the diagnostic yield and is recommended in cases where there is continued concern for abuse after the initial survey or when results of the initial skeletal survey are unclear or questionable. (See "Orthopedic aspects of child abuse", section on 'Skeletal survey'.)

The skeletal survey and its comparison with other forms of bone imaging (radionuclide bone scan, CT bone scan, whole-body magnetic resonance imaging [MRI], or positron emission tomography) in the detection of fractures are discussed in detail separately. (See "Orthopedic aspects of child abuse", section on 'Radiographic evaluation'.)

Neurologic injury — Central nervous system injuries may occur in children who are shaken violently or who are injured by high-energy forces that are associated with impact. (See "Child abuse: Epidemiology, mechanisms, and types of abusive head trauma in infants and children".)

Intracranial injury — The goal of neuroimaging in the acute setting is the rapid detection of treatable conditions (eg, increased intracranial pressure) and to help establish the diagnosis of child abuse. Subsequent evaluation should be directed toward a complete characterization of the extent and age of the injuries. Neuroimaging (CT or MRI) can be used to detect these injuries, to follow their evolution, and to evaluate the development of secondary complications. CT has the advantage of ready availability, speed, and cost, but delivers ionizing radiation to the brain. MRI obviates the radiation risk, but the risk of sedation to the developing brain is being increasingly appreciated. Rapid MRI sequences that may be obtained with neither radiation nor sedation are suggested for hemodynamically and neurologically stable children undergoing evaluation for child abuse as long as they can be performed in a timely manner and promptly interpreted by a pediatric neuroradiologist. (See "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children", section on 'Neuroimaging' and "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children", section on 'Diagnosis'.)

Neuroimaging should be performed in all infants in whom nonaccidental head injury is suspected. We also suggest neuroimaging in children younger than twelve months of age in whom any type of physical abuse is suspected (table 3) [16,39,80,90,91]. Evidence indicates that if neuroimaging is omitted, then skeletal survey and ophthalmologic evaluation alone are insufficient for discovering occult neurologic injury. For example, in one study, 51 children younger than two years who were admitted to an urban children's hospital with injuries suspicious for child abuse and a normal neurologic examination underwent neuroimaging if they met any of the following "high risk" criteria: rib fractures, multiple fractures, facial injury, or age less than six months [39]. Occult neurologic injury was detected in 19 patients (37 percent, 95% CI 24-50 percent) and would have been missed by skeletal survey alone in five children, all of whom were less than six months of age. Ophthalmologic examination was performed in 14 of the 19 children, but no retinal hemorrhages were detected. In a second study, 29 percent of neurologically normal children under one year old had positive screening neuroimaging. Only one (3 percent) had a positive ophthalmological exam [40]. Among infants less than six months old seen at one children's hospital with externally evident injury, only 52 percent had intracranial imaging, but 38 percent of imaged children had an intracranial hemorrhage identified. Again, the authors could not determine that the external injury was the presenting complaint [86].

Other studies of occult intracranial injury prevalence have had variable results, with imaging yields ranging from a low of 4 to 5 percent to a high of 20 to 29 percent [92-97]. For example, in one study, 10 percent of 363 head CT scans had intracranial injury, but 97 percent of positives were accompanied by clinical indications of head trauma [98]. Even that one child with "occult" head injury had a history of head impact, a scalp injury, and a nonspecific intracranial finding for abuse. The wide variation in results is not easily explained by the studies' inclusion criteria and may be the result of regional variations in abuse prevalence or consultation practices. While some studies found a drop-off in positive return above age six months, others have found persistence of high return, though older children were imaged less often and may thus reflect a selection bias.

Unenhanced CT of the head (with brain and bone windows) is the preferred imaging study for initial evaluation in symptomatic patients with suspected inflicted head trauma. MRI is preferred to head CT in asymptomatic children if there is timely availability of the study and interpretation by a pediatric neuroradiologist. (See "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children", section on 'Imaging'.)

Findings of abuse on neuroimaging include evidence of rotational acceleration/deceleration injury, diffuse subdural hemorrhage, subarachnoid hemorrhage, focal cortical contusions, and/or brain swelling. If the child has suffered impact to the head, contact injuries including skull fractures, scalp hematomas, and cerebral contusions may also be identified. Although head CT has the highest sensitivity and specificity for diagnosing acute hemorrhage, some studies have demonstrated that the initial head CTs of abused infants may appear normal with abnormalities appearing on follow-up scans. Nonetheless, emergency head CTs permit identification of any injury requiring rapid surgical intervention. In addition, certain findings are more consistent with intentional head trauma. Review by an experienced pediatric neuroradiologist is often helpful to diagnose abusive head trauma. (See "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children", section on 'Unstable patients' and "Intracranial subdural hematoma in children: Epidemiology, anatomy, and pathophysiology".)

On the other hand, noncontrast brain MRI is superior in documenting the full extent of intracranial injuries and may demonstrate small extraaxial hemorrhages, parenchymal contusions, evidence of shearing injury, and posterior fossa abnormalities missed by CT. Diffusion-weighted imaging may detect cerebral ischemia not readily perceived on conventional CT and MRI. Even when injuries are identified with both CT and MRI, the enhanced visual clarity of the latter often makes for a better study. Furthermore, MRI is the preferred modality for identifying subacute and chronic injury because it has the highest sensitivity and specificity. However, many critically ill infants may not be able to undergo MRI. In addition, because MRI sometimes fails to identify acute subarachnoid or subdural hemorrhage, it is often obtained three to seven days after injury, rather than at the time of admission. (See "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children", section on 'Stable patients'.)

Skull fractures — All abused children with skull fractures on skeletal survey warrant a head CT to evaluate for brain injury. In patients with severe head trauma or other findings (eg, signs of basilar skull fracture), a head CT may obviate the need for plain radiographs of the skull during skeletal survey, especially if three-dimensional skull reconstruction from the CT is performed [89,99,100]. (See "Skull fractures in children: Clinical manifestations, diagnosis, and management", section on 'Diagnosis and radiologic evaluation' and "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children", section on 'Imaging'.)

Spine injury — Because of the high prevalence of cervical spine injury, especially ligamentous injury, in children with abusive head trauma, patients with intracranial injury on neuroimaging should also undergo cervical spine MRI within two to three days of presentation [101-104]. (See "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children", section on 'Associated injuries'.)

Vertebral compression and spinous process fractures usually are visualized on plain radiographs. Complex fractures should be evaluated by thin-section CT. Fractures associated with clinical findings of spinal cord or nerve root injury (uncommon to rare in abuse) should be evaluated by MRI [105]. (See "Evaluation and acute management of cervical spine injuries in children and adolescents", section on 'Cervical spine imaging'.)

Thoracic injury — Blunt thoracic trauma, whether intentional or unintentional, may result in pulmonary contusion, pneumothorax, pleural effusion, rib fractures, and vascular and tracheobronchial injuries [16,72]. Thus, the radiographic evaluation for thoracic injury sustained in child abuse is the same as that for thoracic injury caused by unintentional trauma [106]. Chest and cervical spine radiographs typically are obtained in the initial assessment. They are followed by CT scan with intravenous (IV) contrast if the patient is stable and internal chest injury is suspected [72]. This evaluation is discussed in detail separately. (See "Thoracic trauma in children: Initial stabilization and evaluation", section on 'Evaluation' and "Thoracic trauma in children: Initial stabilization and evaluation", section on 'Management'.)

Fresh rib fractures may be difficult to detect on plain chest or rib films. A four-view chest radiograph that includes oblique views increases the number of rib fractures detected in possible child abuse victims. However, in some cases, they may become evident only after callus formation develops 10 to 14 days into the healing process. When management decisions cannot await the two-week follow-up study, detection of rib fractures in the acute setting can be improved by other radiological techniques. Traditionally, this has been the role of radionuclide scanning. Some experience with fluorine-18 NaF positron emission tomography and computed tomography of the chest suggest that these methods may also be helpful. (See "Orthopedic aspects of child abuse", section on 'Rib fractures' and "Orthopedic aspects of child abuse", section on 'Skeletal survey'.)

Traumatic esophageal perforation and subsequent formation of an abscess are visualized as prevertebral swelling with subcutaneous emphysema on plain radiographs. Contrast esophagography reveals extravasation of contrast material through the rupture, whereas CT with oral contrast delineates the characteristic findings of extraluminal air and fluid and esophageal thickening, as well as extravasation and abscess formation [107]. Traumatic esophageal perforation is discussed in detail separately. (See "Overview of intrathoracic injuries in children", section on 'Esophageal injury'.)

Pneumothorax and hemothorax may result from abusive injuries and are visible on plain radiographs as air and blood in the pleural space, respectively. (See "Spontaneous pneumothorax in children".)

Abuse-related chylothorax is visualized on chest radiograph as unilateral (commonly left-sided) or bilateral pleural fluid collections [108,109]. Abusive chylothorax may be accompanied by rib fractures, and detailed rib films as well as a complete skeletal survey (table 6) should be performed if chylothorax is thought to be inflicted. Additional findings may include subdural hematoma and retinal hemorrhages [108]. (See "Child abuse: Epidemiology, mechanisms, and types of abusive head trauma in infants and children".)

Abdominal injury — Blunt abdominal trauma, whether intentional or unintentional, may result in injury to the liver, spleen, pancreas, kidney, adrenal glands, mesentery, or intestines. Children who are abused have an increased occurrence of pancreatic injuries and duodenal hematomas, injuries that often are difficult to detect on the basis of clinical or radiologic evaluation. Blunt abdominal injury is discussed in detail separately. (See "Pediatric blunt abdominal trauma: Initial evaluation and stabilization".)

Making the diagnosis of intraabdominal injuries in the absence of a history of trauma can be difficult, particularly when the presentation is delayed, as occurs commonly in abusive injuries. In addition, initial laboratory analysis (complete blood count, electrolytes, liver and pancreatic enzymes) may be normal soon after injury, and physical examination may be relatively nonspecific (eg, diffuse abdominal tenderness with hypoactive bowel sounds) or even normal [110].

In contrast to children with nonabusive injuries, it can be important to identify relatively minor abusive injuries, even if they are self-limited, because such injuries can carry forensic significance. Like rib fractures or metaphyseal fractures that do not require specific treatment, the identification of intraabdominal injury can significantly affect the plausibility of an offered trauma history.

We suggest abdominal CT with IV contrast to evaluate for abusive injuries in children with clinical concern for abdominal injury (abdominal bruising, peritonitis, or a positive abdominal ultrasound), a history of significant direct abdominal trauma (such as a direct blow) with abdominal tenderness, or significantly elevated AST or ALT (>80 IU/L).

CT of the abdomen and pelvis with IV contrast is the primary imaging modalities for the evaluation of intraabdominal injury. CT is both sensitive and specific in diagnosing liver, spleen, and retroperitoneal injuries [111]. (See "Pediatric blunt abdominal trauma: Initial evaluation and stabilization", section on 'Radiologic evaluation'.)

Ophthalmology consultation — Consultation with an ophthalmologist for a complete dilated indirect funduscopic examination is recommended in all children younger than five years of age in whom abusive head trauma is strongly suspected. The examination should take place within 24 to 72 hours and, whenever possible, by 48 hours so that transient retinal changes are not missed. The detection of retinal hemorrhages may be a crucial step in the identification of child abuse and appropriately direct additional evaluation. (See "Child abuse: Eye findings in children with abusive head trauma (AHT)", section on 'Ophthalmology consultation'.)

Retinal findings, such as retinal hemorrhages, may be detected by any physician who is familiar with such hemorrhages and has experience with direct fundoscopy (figure 3). Clinicians other than ophthalmologists may also dilate the pupils to improve their view (table 8). However, detection of intraocular abnormalities is optimized by ophthalmology consultation (see "Child abuse: Eye findings in children with abusive head trauma (AHT)", section on 'Initial eye examination'). While detection of retinal hemorrhage on susceptibility-weighted MRI has been reported, this is not recommended as a substitute for clinical examination by an ophthalmologist [112].

Observational evidence suggests that the frequency of retinal hemorrhages is <1 percent in young children with normal neuroimaging and no evidence of head trauma on physical examination [113,114]. Some experts omit ophthalmology examination in these patients, especially if the skeletal survey does not suggest child abuse.

DIAGNOSIS — Once the initial medical, laboratory, and radiologic evaluations are complete and the child's urgent medical needs are addressed, the evaluating physician must determine the level of suspicion for child abuse (ie, whether the injuries appear to have been inflicted or unintentionally acquired) [115]. A reasonable medical suspicion that an injury was inflicted is sufficient to refer the case to social service and law enforcement agencies for further investigation. Examples of such injuries by type and age are provided in the table (table 2). (See 'Reporting and documenting suspected abuse' below.)

Consultation with a multidisciplinary team (eg, social worker, nurse, and child abuse specialist) should also be obtained when available and whenever further evaluation or management or the need to report is unclear [27,115]. Child abuse experts can also help with coordinating multi-disciplinary care, estimating the timing of some injuries, or communicating the level of concern for abuse to other professionals.

The accuracy of multidisciplinary expertise for identifying children with possible abuse has been shown to be high and reinforces the importance of their involvement in all cases of suspected child abuse. For example, in one observational study, a blinded panel of child abuse and pediatric emergency medicine specialists and a biomechanical engineer reviewed clinical data on 584 cases with corroborating non-medical evidence for abusive or accidental injury. The panel was greater than 95 percent accurate for identification of abuse and greater than 99 percent accurate for identification of accidental injury [116].

Establishing the diagnosis of abusive head trauma in infants in children is discussed in greater detail separately. (See "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children", section on 'Diagnosis'.)

DIFFERENTIAL DIAGNOSIS — Conditions that may be confused with physical child abuse are discussed in detail separately. (See "Differential diagnosis of suspected child physical abuse".)

The differential diagnosis of suspected child abuse varies depending upon the clinical manifestations. Familiarity with the medical conditions or cultural practices that mimic child abuse can facilitate:

●Arrival at the correct diagnosis

●Initiation of appropriate therapy

●Avoidance of the consequences of an unnecessary evaluation and/or report of suspected child abuse

The presence of multiple types of injuries (eg, bruising and fractures) suggests abuse because conditions that mimic abuse typically only cause one type of finding.

However, the presence of a condition that mimics child abuse does not exclude the possibility that abuse has occurred. The full clinical picture including all historical features and physical findings must be assessed when abuse is suspected.

MANAGEMENT — Management of children with physical child abuse first consists of stabilization, identification, and treatment of all injuries, including hospitalization for serious injuries such as intracranial or intraabdominal injury, burns with large surface area, femur fractures, or multiple traumatic injuries. The seriously injured patient should receive ongoing care directed by a trauma surgeon with pediatric expertise and/or an appropriate pediatric surgical subspecialist whenever possible. (See "Trauma management: Approach to the unstable child", section on 'Definitive care'.)

The child abuse evaluation is ideally performed in consultation with a multidisciplinary child abuse team led by a child abuse specialist to aid in case management and communication of findings to the parents/caregivers and Child Protective Services.

Reporting and documenting suspected abuse — Mandatory reporting of a suspicion of abuse is required for physicians and other medical providers in many regions. Clinicians must know and abide by mandatory reporting statutes in the jurisdictions where they practice. (See "Child abuse: Social and medicolegal issues", section on 'Reporting suspected abuse'.)

Documenting each step of the evaluation and management of suspected child abuse is of utmost importance. The record of the history should be factual. Statements from the child or parents/caregivers should be recorded as direct quotations to lessen the chance that they will be considered "hearsay." Injuries should be described in as much detail as possible. Sketches and/or high-quality photographs are helpful in documenting extensive injuries. However, in some states, parental permission must be obtained before photographs may be taken. Proper documentation of child abuse injuries is discussed in more detail separately. (See "Child abuse: Social and medicolegal issues", section on 'The written report'.)

Disposition — Children with serious injuries (eg, intracranial or intraabdominal injury, femur fracture, or extensive burns) should be admitted to a pediatric trauma center, whenever possible, with management of their acute injuries directed by a pediatric surgeon and/or appropriate pediatric subspecialist. (See "Trauma management: Approach to the unstable child", section on 'Definitive care'.)

In addition, hospitalization may be indicated when the ongoing safety of the child remains a concern. In this situation, hospitalization permits continued contact between the child and a possibly nonabusive caretaker while the assessment is being completed rather than removal from the parents/primary caregivers [27,117].

Children with injuries that do not require hospitalization may be discharged when the abuse evaluation is complete and a safe outpatient setting has been identified in conjunction with Child Protective Services [27,117].

Identifying injuries in other children at risk — Identifying and ensuring evaluation of all other children currently sharing a household with an abused child is an essential component of case management because of the significant risk of abuse [118-121]. For example, in a prospective study of 520 sibling pairs, when one sibling was determined to be physically abused (7.1 percent of all sibling pairs), over one-half of siblings also had a maltreatment report (relative risk [RR] 16.8, 95% CI 9.7-29.0) and maltreatment was substantiated in 29.7 percent (RR 26.4, 95% CI 12.2-57.6) [118,119]. When the index case is severe or results in death, child maltreatment among other children in the household has been reported in up to 72 percent of siblings [120,121]. In many settings, the ultimate responsibility for this task of identifying and evaluating household contacts falls to the governmental authority responsible for child abuse investigation in collaboration with a child abuse team. However, these children sometimes are brought to the emergency department by social services or have accompanied the abused child.

Children over the age of two years should undergo a complete history and careful physical examination looking for clinical findings suggesting abuse. A limited number of these patients may need additional studies (eg, plain radiographs) to assess acute injuries.

In addition to history and physical examination, we suggest that children two years of age or younger who are household contacts of an abused child receive a skeletal survey to screen for abusive fractures. In a prospective, multicenter observational study of 479 household contacts of abused children undergoing evaluation by child abuse teams, at least one abusive fracture was found in 12 percent of the 134 contacts who were two years of age or younger, including 34 percent of contacts younger than one year of age and 56 percent of twins. Bruises concerning for abuse were noted in 2 percent of children five years of age and younger [122]. None of the fractures detected by screening were clinically suspected or had associated physical findings. Twin siblings were at a clinically significant increased risk of fracture (odds ratio [OR] 20). In a separate observational study of 19 sibling sets (18 twins and one set of triplets), physical abuse or neglect of one twin or triplet was associated with similar maltreatment in 53 percent of the remaining siblings [123]. A high frequency of fractures and abdominal trauma were identified in these affected siblings. A European report on outcomes of skeletal survey in 296 young children identified only 21 children evaluated as the sibling or twin of another abused child, but 33 percent of these studies were positive for occult fracture [85]. (See 'Skeletal survey' above.)

CT of the head in all household contacts six months of age or younger has been proposed [122]. However, evidence is limited regarding the yield of this approach and whether the radiation exposure is justified. Head CT is indicated in these infants and young children despite a normal neurologic examination if they have abusive fractures, retinal hemorrhages, or other physical findings of abuse. (See "Child abuse: Evaluation and diagnosis of abusive head trauma in infants and children", section on 'Unstable patients' and "Physical child abuse: Recognition", section on 'Red flag physical findings'.)

SCREENING — Screening for child abuse is discussed separately. (See "Physical child abuse: Recognition", section on 'Screening in primary care settings'.)

ADDITIONAL RESOURCES — Resources that may be helpful in the evaluation and management of suspected child abuse are listed in the provided table (table 9A-B).

Regional poison control centers — Regional poison control centers in the United States are available at all times for consultation on patients with known or suspected poisoning, and who may be critically ill, require admission, or have clinical pictures that are unclear (1-800-222-1222). In addition, some hospitals have medical toxicologists available for bedside consultation. Whenever available, these are invaluable resources to help in the diagnosis and management of ingestions or overdoses. Contact information for poison centers around the world is provided separately. (See "Society guideline links: Regional poison control centers".)

Society guideline links — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Child abuse and neglect".)

SUMMARY AND RECOMMENDATIONS

●Rationale – The identification of suspected abuse is urgently required not only to treat the current condition, but also to protect the child from a subsequent, perhaps more serious injury. An abused child may have as much as a 50 percent chance of incurring further abuse and a 10 percent chance of dying if abuse is not detected at the initial presentation. (See 'Rationale' above.)

●Red flag history – Historical factors that raise the suspicion of abuse include (see 'History' above):

•The history provided by the parent or caretaker is inconsistent with the injuries of the child (eg, a child with a history of falling from a sofa onto a carpeted floor who presents with multiple fractures, serious intraabdominal injury, or severe head trauma).

•The history is vague or lacking in detail.

•The history changes in repeated versions given by the same caretaker to different health care workers, or conflicting histories are given by different family members/witnesses (eg, infant was burned while bathing in the sink/infant was burned while bathing in the bathtub; mother was bathing infant/sister was bathing infant).

•The injury is attributed to actions of siblings; this account may be a cover story or may be true and related to sibling rivalry, inadequate supervision, or violence in the home. (See "Child neglect: Evaluation and management" and "Intimate partner violence: Childhood exposure".)

•No history is offered (eg, "I don't know what happened; their leg was just suddenly broken").

•The history is inconsistent with the developmental stage of the child (eg, a four-month-old infant who turns on the hot water in the bathtub and is burned).

•The history is implausible (eg, a child presents with swelling of the thigh, with a history given by the parents/caregivers that "a roach bit them"; radiographs reveal a spiral fracture of the femur).

•There is a history of prior bruising or orofacial injury in an infant who is not cruising.

●Red flag physical findings – Findings that raise the suspicion of child abuse include (table 2) (see 'Physical examination' above):

•Injuries that are not consistent with the history, especially those with patterns that indicate a method of infliction

•Injuries commonly associated with abuse

•Multiple fractures in various stages of healing or different types of injuries coexisting (eg, bruises, burns, and/or fractures)

●Laboratory studies and imaging – The laboratory and radiologic evaluation should be tailored to patient age and the clinical findings (table 3). (See 'Initial physical abuse evaluation' above and 'Imaging' above.)

●Management and reporting – After complete evaluation and treatment of the child's urgent medical needs, the evaluating clinician should determine the level of suspicion for child abuse (ie, whether the injuries appear to have been inflicted or unintentionally acquired). Whenever possible, this determination should be made in consultation with a multidisciplinary child abuse team (eg, social worker, nurse, and child abuse specialist). (See 'Management' above.)

In many parts of the world, a report to appropriate governmental authorities is also required for cases of suspected abuse. Reporting of child abuse and neglect is discussed in detail separately. (See "Child abuse: Social and medicolegal issues".)

Hospitalization may also be warranted for the treatment of certain conditions (eg, burns, ingestion, or intracerebral injury) and may also be indicated for the safety of the child. In addition, children currently sharing a household with an abused child require prompt evaluation for possible abuse. (See 'Disposition' above and 'Identifying injuries in other children at risk' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Erin Endom, MD, who contributed to earlier versions of this topic review.