Vous êtes sur la page 1sur 15

Evaluating Children With Fractures for Child Physical Abuse

Emalee G. Flaherty, Jeannette M. Perez-Rossello, Michael A. Levine, William L.

Pediatrics 2014;133;e477; originally published online January 27, 2014;
DOI: 10.1542/peds.2013-3793

The online version of this article, along with updated information and services, is
located on the World Wide Web at:

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned,
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
Boulevard, Elk Grove Village, Illinois, 60007. Copyright 2014 by the American Academy
of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org by guest on February 2, 2014

Guidance for the Clinician in
Rendering Pediatric Care


Evaluating Children With Fractures for Child Physical

Emalee G. Flaherty, MD, Jeannette M. Perez-Rossello, MD,
Michael A. Levine, MD, William L. Hennrikus, MD, and the abstract
Fractures are common injuries caused by child abuse. Although the
ENDOCRINOLOGY, and SECTION ON ORTHOPAEDICS, and the consequences of failing to diagnose an abusive injury in a child
SOCIETY FOR PEDIATRIC RADIOLOGY can be grave, incorrectly diagnosing child abuse in a child whose frac-
KEY WORD tures have another etiology can be distressing for a family. The aim of
fractures this report is to review recent advances in the understanding of frac-
ture specicity, the mechanism of fractures, and other medical dis-
CMLclassic metaphyseal lesions eases that predispose to fractures in infants and children. This
CPRcardiopulmonary resuscitation clinical report will aid physicians in developing an evidence-based dif-
CTcomputed tomography ferential diagnosis and performing the appropriate evaluation when
OIosteogenesis imperfect
assessing a child with fractures. Pediatrics 2014;133:e477e489
This document is copyrighted and is property of the American
Academy of Pediatrics and its Board of Directors. All authors
have led conict of interest statements with the American
Academy of Pediatrics. Any conicts have been resolved through INTRODUCTION
a process approved by the Board of Directors. The American
Academy of Pediatrics has neither solicited nor accepted any Fractures are the second most common injury caused by child physical
commercial involvement in the development of the content of
abuse; bruises are the most common injury.1 Failure to identify an
this publication.
injury caused by child abuse and to intervene appropriately may
The guidance in this report does not indicate an exclusive
course of treatment or serve as a standard of medical care. place a child at risk for further abuse, with potentially permanent
Variations, taking into account individual circumstances, may be consequences for the child.24 Physical abuse may not be considered
appropriate. in the physicians differential diagnosis of childhood injury because
the caregiver may have intentionally altered the history to conceal the
abuse.5 As a result, when fractures are initially evaluated, a diagnosis
of child abuse may be missed.3 In children younger than 3 years, as
many as 20% of fractures caused by abuse may be misdiagnosed
initially as noninicted or as attributable to other causes.3 In addition,
fractures may be missed because radiography is performed before
changes are obvious or the radiographic images are misread or
misinterpreted.2 However, incorrectly diagnosing physical abuse in a child
with noninicted fractures has serious consequences for the child and
www.pediatrics.org/cgi/doi/10.1542/peds.2013-3793 family. To identify child abuse as the cause of fractures, the physician
doi:10.1542/peds.2013-3793 must take into consideration the history, the age of the child, the location
All clinical reports from the American Academy of Pediatrics and type of fracture, the mechanism that causes the particular type of
automatically expire 5 years after publication unless reafrmed, fracture, and the presence of other injuries while also considering other
revised, or retired at or before that time.
possible causes.
PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).
Copyright 2014 by the American Academy of Pediatrics
Trauma: Child Abuse Versus Noninicted Injuries
Fractures are a common childhood injury and account for between 8%
and 12% of all pediatric injuries.68 In infants and toddlers, physical

PEDIATRICS Volume 133, Number 2, February 2014 e477

Downloaded from pediatrics.aappublications.org by guest on February 2, 2014
abuse is the cause of 12% to 20% of Fracture Specicity for Abuse in infants who have increased bone
fractures.9 Although unintentional fragility.2325
Fractures With High Specicity for
fractures are much more common Cardiopulmonary resuscitation (CPR)
than fractures caused by child abuse, has been proposed as a cause of rib
the physician needs to remain aware As shown in Table 1, certain fractures
fractures, but conventional CPR with 2
of the possibility of inicted injury. have high specicity for or strong as-
ngers of 1 hand rarely causes frac-
Although some fracture types are sociation with child abuse, particularly
in infants, whereas others may have tures in children.26,27 Recent recom-
highly suggestive of physical abuse, mendations that CPR be performed
no pattern can exclude child abuse.10,11 less specicity.21 Rib fractures in
infants, especially those situated post- using 2 hands encircling the rib cage
Specically, it is important to recog- have raised concerns that this tech-
nize that any fracture, even fractures eromedially, and the classic meta-
physeal lesions of long bones, have nique might cause rib fractures. An
that are commonly noninicted inju- analysis of infants who were discov-
ries, can be caused by child abuse. high specicity for child abuse. Frac-
tures of the scapula, spinous process, ered during autopsy to have rib frac-
Certain details that can help the tures and had received 2-handed
physician determine whether a frac- and sternum also have high specicity
for abuse but are uncommon. chest compressions antemortem
ture was caused by abuse rather suggested that 2-handed CPR is as-
than unintentional injury include the Rib fractures are highly suggestive of
sociated with anterior-lateral rib
history, the childs age and de- child abuse. Most abusive rib fractures
fractures of the third to sixth ribs.28
velopmental stage, the type and lo- result from anterior-posterior com-
In this small study, no posterior rib
cation of the fracture, the age of the pression of the chest. For this reason,
fractures were observed. The frac-
fracture, and an understanding of the rib fractures are frequently found in
tures in these infants were always
mechanism that causes the particu- infants who are held around the chest,
multiple, uniformly involved the
lar type of fracture. The presence of squeezed, and shaken. Rib fractures
fourth rib, and were sometimes bi-
multiple fractures, fractures of dif- have high probability of being caused
lateral. Additional research is needed
ferent ages or stages of healing, de- by abuse.15,17,21 The positive predictive
to examine the relationship between
lay in obtaining medical treatment, value of rib fractures for child abuse in
the 2-handed CPR technique and rib
and the presence of other injuries children younger than 3 years was
suspicious for abuse (eg, coexisting 95% in one retrospective study.22 Other
injuries to the skin, internal organs, less common causes of rib fractures in Classic metaphyseal lesions (CMLs)
or central nervous system) should infants include signicant trauma also have high specicity for child
alert the physician to possible child sustained during childbirth or a motor abuse when they occur during the
abuse. vehicle crash as well as minor trauma rst year of life.21,29 CMLs are the
most common long bone fracture
found in infants who die with evi-
Childs Age and Development TABLE 1 Specicity of radiologic ndings in dence of inicted injury.30 CMLs are
infants and toddlers19 planar fractures through the pri-
The physician should consider the
childs age and level of development. High specicitya mary spongiosa of the metaphysis.
Approximately 80% of all fractures CMLs These fractures are caused when
Rib fractures, especially posteromedial torsional and tractional shearing
caused by child abuse occur in chil- Scapular fractures
dren younger than 18 months,12 and Spinous process fractures
strains are applied across the met-
approximately one-quarter of fractures Sternal fractures aphysis, as may occur with vigorous
Moderate specicity pulling or twisting of an infants ex-
in children younger than 1 year are Multiple fractures, especially bilateral
caused by child abuse.1,9,1315 Physical Fractures of different ages
tremity.31 Fractures resembling
abuse is more likely to be the cause of Epiphyseal separations CMLs radiographically have been
femoral fractures and humeral frac- Vertebral body fractures and subluxations reported after breech delivery32 and
Digital fractures
tures in children who are not yet as a result of treatment of club-
Complex skull fractures
walking compared with children who Common, but low specicity foot.33
are ambulatory,1518 and the percent- Subperiosteal new bone formation Depending on the projection of the
Clavicular fractures
age of fractures caused by abuse Long-bone shaft fractures radiograph, CMLs can have the ap-
declines sharply after the child begins Linear skull fractures pearance of a corner or a bucket-
to walk.9,19,20 a
Highest specicity applies in infants. handle fracture. Acute injuries can


Downloaded from pediatrics.aappublications.org by guest on February 2, 2014

be difcult to visualize radiographi- to the bone, whereas spiral fractures down several steps and landing with 1
cally. CMLs commonly heal without are caused by torsion or twisting of leg folded or twisted underneath
subperiosteal new bone formation or a long bone along its long axis. Obli- a child can lead to excessive torsional
marginal sclerosis. They can heal que fractures are caused by a combi- loading of the femur and a spiral
quickly and be undetectable on plain nation of bending and torsion loads.37 fracture.46 In ambulatory children,
radiographs in 4 to 8 weeks.31 Torus or buckle fractures are the re- noninicted femoral fractures have
sult of compression from axial loading been described in children who fell
Fractures With Moderate Specicity along the length of the bone. Although while running or who fell and landed
for Abuse earlier studies suggested that spiral in a split-leg position.43
Although many children who have fractures should always raise suspi- A fracture of the humeral shaft in a child
been abused will have only a single cion for child abuse,12 more recent younger than 18 months has a high
fracture,34 the presence of multiple studies do not show that any partic- likelihood of having been caused by
fractures, fractures of different ages ular fracture pattern can distinguish abuse.15,49,50 In contrast, supracondylar
and/or stages of healing, and com- between abuse and nonabuse with fractures in ambulatory children are
plex skull fractures have moderate absolute certainty.16,38 usually noninicted injuries resulting
specicity for physical abuse. In ad- Falls are common in childhood.39 Short from short falls.15
dition, epiphyseal separations, verte- falls can cause fractures, but they Physicians should also be aware of
bral body fractures, and digital rarely result in additional signicant a particular mechanism reported to
fractures have moderate specicity injury (eg, neurologic injury).11,4042 In produce a noninicted spiral-oblique
for abuse. The presence of multiple a retrospective study of short falls, fracture of the humerus in 1 case
fractures or fractures of different parents reported that 40% of the report.51 When the young infant was
ages can be signs of bone fragility children before 2 years of age had rolled from the prone position to the
but should also evoke consideration suffered at least 1 fall from a height of supine while the childs arm is ex-
of child abuse. Besides the predictive between 6 inches and 4 feet. Approx- tended, the torsion and stress placed
value of the particular pattern of imately one-quarter of these children on the extended arm appeared to
fractures, many other factors, such suffered an injury; bruises were the cause a spiral-oblique fracture of the
as the history and the childs age, most common injury observed.43 midshaft of the humerus.
must be considered when de- The femur, humerus, and tibia are the Linear skull fractures of the parietal
termining whether the injury was most common long bones to be in- bone are the most common skull
inicted. jured by child abuse.1,34 Femoral fracture among young children, usu-
fractures in the nonambulatory child ally children younger than 1 year.13 A
Common Fractures With Low are more likely caused by child short fall from several feet onto
Specicity for Child Abuse abuse, whereas these fractures in a hard surface can cause a linear,
Long bone fractures (other than CMLs), ambulatory children are most com- nondiastatic skull fracture.19,52 The
linear skull fractures, clavicle fractures, monly noninicted.10,16,4345 majority of linear skull fractures are
and isolated ndings of subperiosteal Certain femur fractures may occur as not inicted.53 By contrast, complex or
new bone formation have low speci- a result of a noninicted injury in young bilateral skull fractures are typical of
city for child abuse. In contrast, the children. Several studies have demon- nonaccidental trauma.
single long bone diaphyseal fracture is strated that a short fall to the knee may
the most common fracture pattern produce a torus or impacted transverse Syndromes, Metabolic Disorders,
identied in abused children.1,13,34 fracture of the distal femoral meta- Systemic Disease
An understanding of the extent and diaphysis.46,47 Oblique distal femur met- Preexisting medical conditions and
type of load that is necessary to cause aphyseal fractures have been reported bone disease may make a childs
a particular long bone fracture can in children playing in a stationary ac- bones more vulnerable to fracture.
help to determine whether a specic tivity center, such as an Exersaucer Some conditions may manifest skele-
fracture is consistent with the injury (Eveno, Picqua, OH).48 tal changes, such as metaphyseal ir-
described by the caregiver.35,36 Trans- In both ambulatory and nonambulatory regularity and subperiosteal new
verse fractures of the long bones are children, under some circumstances, bone formation. These entities should
caused by the application of a bending falls on a stairway can cause a spiral be considered in the differential di-
load in a direction that is perpendicular femoral fracture. For example, a fall agnosis of childhood fractures.

PEDIATRICS Volume 133, Number 2, February 2014 e479

Downloaded from pediatrics.aappublications.org by guest on February 2, 2014
Osteogenesis Imperfecta long bones. It is unusual to have mul- Although osteopenia of prematurity
Osteogenesis imperfecta (OI) is a het- tiple long bone fractures or rib frac- may make the infant more vulnerable
erogeneous family of diseases, usually tures, particularly in infancy, without to fracture, preterm infants are also at
caused by heterozygous mutations of other clinical and radiographic evi- an increased risk of abuse.68
the genes COL1A1 and COL1A2,54 but dence of OI.57,58
mutations in these and other genes OI has been misdiagnosed as child Vitamin D Deciency Rickets
can cause autosomal recessive forms abuse.59 On the other hand, OI is often
Suboptimal vitamin D concentrations
of OI. The COL1A1 and COL1A2 genes suggested as the cause of fractures in
and rickets have been proposed as
encode the chains of type I collagen, children who have been abused. If
causes of fractures in infants.69 Vitamin
which forms the structural framework fractures continue to occur when
D insufciency in otherwise healthy
of bone. Although it is a genetic dis- a child is placed in a protective envi-
infants and toddlers is common. Ap-
order, many children have de novo ronment, a more thorough evaluation
proximately 40% of infants and tod-
mutations or autosomal-recessive dis- for an underlying bone disease is
dlers aged 8 to 24 months in an urban
ease and no family history of bone needed. Child abuse is more common
clinic had laboratory evidence of vita-
fragility. In addition, the presentation of than OI,60 and children with OI and
min D insufciency (serum concen-
the disease within affected members other metabolic or genetic conditions
trations of 25-hydroxyvitamin D of 30
of the same family can be quite vari- may also be abused.61,62
ng/mL).70 Prolonged breastfeeding
able. Phenotypic expression of the
without vitamin D supplementation
disease depends on the nature of the Preterm Birth was a critical factor that placed these
mutation, its relative abundance at-
Preterm infants have decreased bone infants at risk, although increased skin
tributable to mosaicism, and its ex-
mineralization at birth, but after the pigmentation and/or lack of sunlight
pression in target tissues.55 Some
rst year of life, bone density nor- exposure may also have contributed.
types of OI involve reduced production
malizes.63,64 Osteopenia of prematurity Rickets is characterized by de-
of collagen, and the symptoms resolve
has been well described as a compli- mineralization, loss of the zone of
or lessen after puberty.56 Table 2 lists
cation in low birth weight infants.65 provisional calcication, widening
the various signs and symptoms that
Infants born at less than 28 weeks and irregularity of the physis, and
can be present in a case of OI.
gestation or who weigh less than 1500 fraying and cupping of the meta-
The diagnosis of OI is often suggested g at birth are particularly vulnerable. physis.71 Despite the high prevalence
by a family history of fractures, short Osteopenia of prematurity is multi- of vitamin D insufciency in infants
stature, blue sclera, poor dentition, factorial. Infants are also at risk if and toddlers, rickets is uncommon.72
and radiographic evidence of low bone they receive prolonged (for 4 or more
density or osteopenia. The fractures weeks) total parenteral nutrition, have The claim that vitamin D deciency or
are most commonly transverse in bronchopulmonary dysplasia, and/or insufciency causes skeletal lesions
nature, occurring in the shafts of the have received a prolonged course of that lead to the incorrect diagnosis of
diuretics or steroids.66 Osteopenia child abuse in infants is not sup-
TABLE 2 Characteristics of Osteogenesis
commonly presents between 6 and 12 ported in the literature. A systematic
Imperfecta weeks of life. Osteopenia of pre- clinical, laboratory, and radiologic
maturity can be ameliorated if infants assessment should exclude that
Fragile bones with few, some, or many of the
following ndings: are monitored closely and receive the possibility.7375 Schilling et al found
Poor linear growth nutritional and mineral supplementa- no difference in serum concen-
Macrocephaly tion initiated in the NICU.
Triangular-shaped face
trations of 25-hydroxyvitamin D in
Blue sclerae Fractures associated with osteopenia of young children with fractures suspi-
Hearing impairment as a result of otosclerosis prematurity usually occur in the rst cious for abuse and noninicted
Hypoplastic, translucent, carious, late-erupting,
or discolored teeth
year of life.67 Rib fractures are typically fractures.76 Vitamin D insufciency
Easy bruisability encountered incidentally, whereas long was not associated with multiple
Inguinal and/or umbilical hernias bone fractures commonly present with fractures, in particular rib fractures
Limb deformities swelling of the extremity. Osteopenia of or CMLs, the high specicity indica-
Hyperextensible joints
Scoliosis and/or kyphosis prematurity can be associated with tors of abuse. Perez-Rossello et al
Wormian bones of the skull rickets, and in such cases, meta- studied radiographs of 40 healthy
Demineralized bones physeal irregularities may be present. older infants and toddlers with


Downloaded from pediatrics.aappublications.org by guest on February 2, 2014

vitamin D insufciency and deciency inicted and noninicted fractures oc- months of age, because fetal copper
and concluded that radiographic ra- curring in these children. At the same stores are sufcient for this length
chitic changes were uncommon and time, children with disabilities are at an of time. In addition, human milk and
very mild. In this population, the increased risk of being maltreated.8284 formula contain sufcient copper to
reported fracture prevalence was When multiple or recurrent fractures prevent deciency. Psychomotor re-
zero.72 occur in a disabled child, a trial change tardation, hypotonia, hypopigmentation,
In a study of 45 young children with in caregivers may be indicated to de- pallor, and a sideroblastic anemia are
radiographic evidence of rickets, termine whether the fractures can be some of the characteristic ndings of
investigators found that fractures oc- prevented. This is an extreme in- copper deciency in infants. Radiologic
curred only in those infants and tod- tervention and should be reserved for changes that should lead to further
dlers who were mobile.77 Fractures unusual circumstances.63 evaluation for possible deciency
were seen in 17.5% of the children, include cupping and fraying of the
and these children were 8 to 19 Scurvy metaphyses, sickle-shaped metaphyseal
months of age. The fractures involved spurs, signicant demineralization, and
Scurvy is caused by insufcient intake
long bones, anterior-lateral and lat- subperiosteal new bone formation.
of vitamin C, which is important for the
eral ribs, and metatarsal and meta- synthesis of collagen. Although rare
physeal regions. The metaphyseal today because formula, human milk, Menkes Disease
fractures occurred closer to the di- fruits, and vegetables contain vitamin Menkes disease, also known as
aphysis in the background of orid C, scurvy may develop in older infants Menkes kinky hair syndrome, is a rare
metaphyseal rachitic changes and and children given exclusively cow congenital defect of copper metabo-
did not resemble the juxtaphyseal milk without vitamin supplementation lism.90 Menkes disease is an X-linked
corner or bucket handle pattern of and in children who eat no foods recessive condition and occurs only in
the CML. In infant fatalities in which containing vitamin C.8587 Although boys. Although it has many of the
abuse is suspected, rachitic changes scurvy can result in metaphyseal features of dietary copper deciency,
appear to be rare histologically.78 changes similar to those seen with anemia is not associated with Menkes
child abuse, other characteristic bone disease. Metaphyseal fragmentation
Osteomyelitis changes, including osteopenia, in- and subperiosteal new bone forma-
Osteomyelitis in infants can present as creased sclerosis of the zones of tion may be observed on radiographs,
multiple metaphyseal irregularities provisional calcication, dense epiph- and the ndings may be difcult to
potentially resembling CMLs.79 Typically, yseal rings, and extensive calcication distinguish from fractures caused by
the lesions become progressively lytic of subperiosteal and soft tissue hem- abuse.91 Other signs of Menkes dis-
and sclerotic with substantial sub- orrhages, will point to the diagnosis ease include sparse, kinky hair, cal-
periosteal new bone formation. Other of scurvy. varial wormian bones, anterior rib
signs of infection are often present, aring, failure to thrive, and de-
such as fever, increased erythrocyte Copper Deciency velopmental delay. A characteristic
sedimentation rate, elevated C-reactive nding is tortuous cerebral vessels.
Copper plays a role in cartilage for-
protein concentration, and elevated Intracranial hemorrhage can occur in
mation. Copper deciency is a rare
white blood cell count. Menkes disease but has not been
condition that may be complicated by
reported in infants with copper de-
bone fractures. Preterm infants are
Fractures Secondary to ciency.
born with lower stores of copper than
Demineralization From Disuse term infants, because copper is ac-
Any child with a severe disability that cumulated at a faster rate during the Systemic Disease
limits or prevents ambulation can be at last trimester.88 Copper insufciency Chronic renal disease affects bone
risk for fractures secondary to disuse may be observed in children with metabolism because children with
demineralization, even with normal severe nutritional disorders, for ex- chronic renal disease may develop
handling.80,81 The fractures are usually ample, liver failure or short gut syn- a metabolic acidosis that interferes
diaphyseal rather than CMLs. Often, drome.89 This deciency is not likely with vitamin D metabolism. Chronic
these fractures occur during physical to be observed in full-term children renal disease can cause renal osteo-
therapy and range-of-motion exercises. younger than 6 months of age or dystrophy resulting in the same ra-
It can be difcult to distinguish between preterm infants younger than 2.5 diographic changes as nutritional

PEDIATRICS Volume 133, Number 2, February 2014 e481

Downloaded from pediatrics.aappublications.org by guest on February 2, 2014
rickets. Because chronic liver disease for infants. He relied on the mothers some swelling, pain, or other signs,
(eg, biliary atresia) interferes with vi- history of decreased fetal movements such as decreased use of the ex-
tamin D metabolism, such children may and provided no independent mea- tremity, suggesting a fracture.100,101
be at an increased risk of fractures. surements of those movements. Pala- Some children, however, will have
Fanconi syndrome, hypophosphatasia, cios and Rodriguez found no evidence minimal external signs of injury.102
hypophosphatemic (vitamin D resistant) that oligohydramnios affects bone mass The absence of any history of injury,
rickets, hyperparathyroidism, and renal of the fetus, probably because fetal a vague description of the event,
tubular acidosis also cause clinical movement is only restricted in the last a delay in seeking care, the absence
variants of rickets. trimester of pregnancy by oligohy- of an explanation for an injury par-
dramnios and because the mechanical ticularly in a nonambulatory child, or
loading on the bones stimulating bone an inconsistent explanation should
Temporary Brittle Bone Disease
formation is conserved.99 increase the physicians concern that
an injury was caused by child abuse
Physicians should be aware of alter- (see Table 3).13,16
native diagnoses that are unsupported Medical Evaluation
by research but are sometimes sug- Past Medical History
History of Present Illness
gested when an infant has unexplained The past medical history is important
fractures. In 1993, Paterson proposed It is essential to obtain a detailed his-
tory to determine how an injury oc- and should include details about the
that some infants may be born with mothers pregnancy. If the child was
bones that are temporarily more curred. If an injury in a nonverbal child
was witnessed, the caregiver should be born preterm, the infants bone min-
fragile or vulnerable to fracture in the eral content may be reduced, and the
context of normal handling, which he able to provide details about the childs
infant may be at risk for fracture. A
called temporary brittle bone dis- activity and position before an injury
history of total parental nutrition,
ease.92 Paterson suggested that and the childs nal position and loca-
tion after the injury occurred.46 Verbal hepatobiliary disease, diuretic ther-
some trace element deciency, such apy, hypercalciuria, or corticosteroids
as copper or a transient collagen im- children with concerning fractures
should be interviewed apart from may make the bones of a low birth
maturity, caused the disease but pro- weight infant even more vulnerable to
vided no scientic data that conrmed caregivers and ideally by a professional
who is skilled in forensic interviewing. fracture. In addition, chronic diseases,
his hypotheses and offered no specic such as renal insufciency or meta-
test that conrmed temporary brittle A comparison of the histories provided bolic acidosis, malabsorption, cere-
bone disease.61 Subsequent studies did by caregivers of children with non- bral palsy or other neuromuscular
not support his hypotheses, and his inicted femoral fractures and by disorders, genetic diseases that affect
case analysis has been refuted.57,9395 caregivers of children whose injuries skeletal development, or any illness
Miller hypothesized that temporary were caused by abuse is instructive. that limits mobility, may affect bone
brittle bone disease is a result of fetal When an injury was caused by abuse, strength. A thorough dietary history
immobilization or intrauterine con- the caregiver provided either no his- and history of medications that can
nement that leads to transient bone tory of an injury or related a history of
loss or osteopenia.96,97 In support of a low-energy event. By contrast, 29% of
his hypothesis, he reported that 95% the caregivers of children with non- TABLE 3 When Is a Fracture Suspicious for
Child Abuse?
of 21 infants with multiple unexplained inicted injuries provided some high-
fractures had decreased fetal move- energy explanation, such as a motor No history of injury
vehicle collision or that the child fell History of injury not plausiblemechanism
ments, according to their mothers.97,98 described not consistent with the type of
Although he used bone densitometry in from a height.16 Most of the low- fracture, the energy load needed to cause the
each patient as a basis for his con- energy mechanisms provided for the fracture, or the severity of the injury
noninicted injuries involved falls in- Inconsistent histories or changing histories
clusions, none of the patients had had provided by caregiver
bone densitometry performed at the cluding stair falls and siblings landing
Fracture in a nonambulatory child
time of the fracture. The testing was on the femur during play.16,46 Fracture of high specicity for child abuse (eg,
performed 8 to 21 weeks later, and no The childs response to the event may rib fractures)
Multiple fractures
infants were tested before 5 months of also provide important clues about Fractures of different ages
age. In addition, bone densitometry the etiology. The majority of children Other injuries suspicious for child abuse
standards have not been established with long bone fractures will have Delay in seeking care for an injury


Downloaded from pediatrics.aappublications.org by guest on February 2, 2014

predispose to fractures are impor- neck, or trunk should raise suspicion type I collagen and may identify the
tant. The physician should inquire for child abuse.105,106 The child should mutation to guide testing of other
about previous injuries including be examined for other injuries caused family members.107 Some of the less
bruises and determine the childs de- by child abuse, in addition to signs of common forms of OI are OI types IIB
velopmental abilities, because chil- other medical conditions associated and VII, CRTAP; OI type VI, FKBP10; OI
dren who are not yet mobile are much with bone fragility. Blue sclerae are type VIII, LEPRE1; OI type IX, PPIB; OI type
more likely to have fractures caused seen in certain types of OI. Sparse, X, SERPINH1; OI type XI, SP7; OI type XII,
by abuse. kinky hair is associated with Menkes SERPINF1; and OI type XIII, BMP1. DNA
disease. Dentinogenesis imperfecta is sequencing can be performed using
Family History occasionally identied in older chil- genomic DNA isolated from peripheral
A family history of multiple fractures, dren with OI. blood mononuclear cells or even sa-
early-onset hearing loss, abnormally liva, whereas the biochemical analysis
developed dentition, blue sclera, and of type I collagen requires a skin bi-
Laboratory Evaluation
short stature should suggest the opsy. Doing both DNA analysis and
The clinical evaluation should guide skin biopsy is not indicated in most
possibility of OI.
the laboratory evaluation. In children cases. Consultation with a pediatric
with fractures suspicious for abuse, geneticist may be helpful in deciding
Social History serum calcium, phosphorus, and al- which children to test and which test
The physician should obtain a com- kaline phosphatase should be reviewed, to order.108
plete psychosocial history, including although alkaline phosphatase may be
asking who lives in the home and who elevated with healing fractures. The
has provided care for the child. The physician should consider checking Imaging Approach
history should inquire about intimate serum concentrations of parathyroid Children younger than 2 years with
partner violence, substance abuse hormone and 25-hydroxyvitamin D, as fractures suspicious for child abuse
including drugs and alcohol, mental well as urinary calcium excretion (eg, should have a radiographic skeletal
illness, and previous involvement with random urinary calcium/creatinine survey to look for other bone injuries
child protective services and/or law ratio) in all young children with or osseous abnormalities.109 Addi-
enforcement. fractures concerning for abuse, but tional fractures are identied in ap-
these levels should certainly be proximately 10% of skeletal surveys,
Physical Examination assessed if there is radiographic ev- with higher yields in infants.110 Skeletal
The child should have a comprehensive idence of osteopenia or metabolic surveys may be appropriate in some
physical examination, and the growth bone disease. Screening for abdomi- children between ages 2 and 5 years,
chart should be carefully reviewed. nal trauma with liver function studies depending on the clinical suspicion
Abnormal weight may suggest neglect as well as amylase and lipase con- of abuse. If specic clinical ndings
or endocrine or metabolic disorders. centration should be performed when indicate an injury at a particular
Any signs or symptoms of fractures, severe or multiple injuries are iden- site, imaging of that area should
such as swelling, limitation of motion, tied. A urinalysis should be per- be obtained regardless of the childs
and point tenderness should be formed to screen for occult blood. age.
documented. The physician should do Serum copper, vitamin C, and ceru- The American College of Radiology has
a complete skin examination to look loplasmin concentrations should be developed specic practice guidelines
for bruises and other skin ndings considered if the child is at risk for for skeletal surveys in children.111
because bruises are the most common scurvy or copper deciency and has Twenty-one images are obtained, in-
injury caused by child abuse. The radiographic ndings that include cluding frontal images of the appen-
majority of children with fractures do metaphyseal abnormalities. dicular skeleton, frontal and lateral
not have bruising associated with the If OI is suspected, sequence analysis views of the axial skeleton, and obli-
fracture; the presence or absence of of the COL1A1 and COL1A2 genes that que views of the chest. Oblique views
such bruising does not help to de- are associated with 90% of cases of of the chest have been shown to in-
termine which fractures are caused by OI as well as other genes associated crease the sensitivity, specicity, and
child abuse.103,104 Bruising in a child with less common autosomal- accuracy of the identication of rib
who is not yet cruising or bruising in recessive forms of OI may be more fractures.112 A full 4 skull series should
unusual locations, such as the ears, sensitive than biochemical tests of be obtained if there are concerns of

PEDIATRICS Volume 133, Number 2, February 2014 e483

Downloaded from pediatrics.aappublications.org by guest on February 2, 2014
head injury. Computed tomography (CT) the specic clinical indicators of of equivocal or negative skeletal sur-
3-dimensional models are valuable abuse.109 veys when there is high clinical suspi-
adjuncts to the radiographs and have Because brain injuries are often occult, cion of abuse. If available, a 18F-NaF
the potential to replace the skull se- head imaging should be considered for positron emission tomography bone
ries.113 This has not been studied sys- any child younger than 1 year with scan has better contrast and spatial
tematically in this context, however. a fracture suspicious for abuse.121 Im- resolution than 99mTc-labeled methy-
Because lateral views of the extremities aging studies may help clarify whether lene diphosphonate.120
increase yield, some authors suggest the child has been abused, provide Although bone densitometry by dual-
that these views be included in the further support for a diagnosis of child energy x-ray absorptiometry is useful
imaging protocol.114 Fractures may be abuse, and identify other injuries that to predict bone fragility and fracture
missed if the guidelines are not fol- require treatment. Additional imaging risk in older adults, interpretation of
lowed or if the images are of poor may be needed if the child has signs or bone densitometry in children and
quality.115 A repeat skeletal survey symptoms of chest, abdominal, or neck adolescents is more problematic.126 In
should be performed approximately 2 injury. adults, bone densitometry is inter-
to 3 weeks after the initial skeletal Chest CT can identify rib fractures that preted using T scores, which describe
survey if child abuse is strongly sus- are not seen on chest radiographs.122 the number of SDs above or below the
pected.109,116 The follow-up examination CT is particularly useful in detecting average peak bone mass for a gender-
may identify fractures not seen on the anterior rib fractures and rib frac- and race-adjusted reference group of
initial skeletal survey, can clarify un- tures at all stages of healingearly normal subjects. Because peak bone
certain ndings identied by the initial subacute, subacute, and old fractures. mass is not achieved until approxi-
skeletal survey, and improves both Although CT may be more sensitive in mately 30 years of age, in children,
sensitivity and specicity of the skeletal identifying these injuries, a chest CT z scores must be used to express bone
survey.116,117 In one study, 13 of 19 exposes the child to signicantly more density, because z scores express the
fractures found on the follow-up ex- radiation than chest radiography. Ev- childs bone mineral density as a func-
amination were not seen on the initial ery effort should be made to reduce tion of SDs above or below the average
series.116 The number of images on the childrens exposure to radiation while for an age- and gender-matched norm
follow-up examination may be limited at the same time considering the risk control population.127 In addition, be-
to 15 views by omitting the views of the to the child if abuse is not identi- cause bone size inuences dual-energy
skull, pelvis, and lateral spine.118 x-ray absorptiometry, z scores must
ed.123 Therefore, selective application
Radiography may assist in assessing also be adjusted for height z scores.128
of this technique in certain clinical
the approximate time when an injury The International Society for Clinical
settings is appropriate.
occurred because long bone fractures Densitometry recommends that the
Other modalities may become available diagnosis of osteoporosis in childhood
heal following a particular sequence.119 in the future that will provide more
If the healing pattern is not consistent should not be made on the basis of low
accurate identication of skeletal inju- bone mass alone but should also in-
with the explanation provided, the ries. Whole-body short tau inverse re-
accuracy of the explanation should be clude a clinically signicant history
covery imaging, a magnetic resonance of low-impact fracture. The recom-
questioned. imaging (MRI) technique, may identify mendations currently apply to chil-
Bone scintigraphy may be used to rib fractures not recognized on the dren 5 years and older, although
complement the skeletal survey but radiographic skeletal survey.124 In reference data are available for chil-
should not be the sole method of a study of 21 infants with suspected dren as young as 3 years.129,130 Un-
identifying fractures in infants. Al- abuse, whole-body MRI at 1.5-Tesla was fortunately, there are limited reference
though it has high overall sensitivity, it insensitive in the detection of CMLs data for the young, nonverbal child
lacks specicity for fracture detection and rib fractures. In some cases, who is most at risk for suffering frac-
and may fail to identify CMLs and skull whole-body MRI identied soft tissue tures caused by child abuse.
fractures.109,119,120 Scintigraphy does edema and joint effusions that led to
have high sensitivity for identifying rib the identication of skeletal injuries
fractures, which can be difcult to with additional radiographs.125 Bone Evaluation of Siblings
detect before healing. In toddlers and scintigraphy with 18F-sodium uoride Siblings, especially twins, and other
older children, the use of bone scin- positron emission tomography (18F-NaF young household members of children
tigraphy or skeletal survey depends on PET) bone scan may be useful in cases who have been physically abused


Downloaded from pediatrics.aappublications.org by guest on February 2, 2014

should be evaluated for maltreat- groups and across all racial and ethnic Emalee G. Flaherty, MD, FAAP
ment.131 In a study of 795 siblings in groups. Many of these diagnoses are John M. Leventhal, MD, FAAP
James L. Lukefahr, MD, FAAP
400 households of a child who had complex. If a physician is uncertain Robert D. Sege MD, PhD, FAAP
been abused or neglected, all sib- about how to evaluate an injury or if
lings in 37% of households and some they should suspect a fracture was LIAISONS
siblings in 20% of households had caused by child abuse, they should Harriet MacMillan, MD American Academy of
suffered some form of maltreat- consult a child abuse pediatrician Child and Adolescent Psychiatry
Catherine M. Nolan, MSW, ACSW Administration
ment.132 In this study, which included or multidisciplinary child abuse team for Children, Youth, and Families
all manifestations of maltreatment, to assist in the evaluation, particu- Linda Anne Valley, PhD Centers for Disease
siblings were found to be more at larly if the child is nonambulatory Control and Prevention
risk for maltreatment if the index or younger than 1 year of age. 134 In
child suffered moderate or severe certain circumstances, the physician STAFF
maltreatment. In addition to a careful will need to consult an orthopedist, Tammy Piazza Hurley
evaluation, imaging should be consid- endocrinologist, geneticist, or other
ered for any siblings younger than 2 subspecialists. SECTION ON RADIOLOGY EXECUTIVE
years, especially if there are signs of COMMITTEE, 20122013
All US states, commonwealths, and
Christopher I. Cassady, MD, FAAP, Chairperson
abuse. territories have mandatory reporting Dorothy I. Bulas, MD, FAAP
requirements for physicians and John A. Cassese, MD, FAAP
DIAGNOSIS other health care providers when Amy R. Mehollin-Ray, MD, FAAP
Maria-Gisela Mercado-Deane, MD, FAAP
When evaluating a child with a frac- child abuse is suspected. Physicians
Sarah Sarvis Milla, MD, FAAP
ture, physicians must take a careful should be aware of and comply with
history of any injury event and the reporting requirements of their
then determine whether the mechan- state. Typically, the standard for Vivian Thorne
ism described and the severity and making a report is when the reporter
timing are consistent with the injury suspects or has reason to believe
identied (see Table 3).133 They must that a child has been abused or EXECUTIVE COMMITTEE, 20122103
consider and evaluate for possible di- neglected. Sometimes determining Irene N. Sills, MD, FAAP, Chairperson
agnoses in addition to other signs or whether that reasonable belief or Clifford A. Bloch, MD, FAAP

symptoms of child abuse. A careful reasonable suspicion standard has Samuel J. Casella, MD, MSc, FAAP
Joyce M. Lee, MD, FAAP
evaluation for other injuries is im- been met can be nuanced and com- Jane Lockwood Lynch, MD, FAAP
portant because the presence of ad- plex. The physician should keep in Kupper A. Wintergerst, MD, FAAP
ditional injuries that are associated mind that incontrovertible proof of
with child abuse increases the likeli- abuse or neglect is not required by STAFF
hood that a particular fracture was state statutes, and there may be Laura Laskosz, MPH
inicted.16,43 It is important to re- cases in which it is reasonable to
member that even if a child has an consult with a child abuse pediatri- SECTION ON ORTHOPEDICS EXECUTIVE
underlying disorder or disability that cian about whether a report should COMMITTEE, 20122013
be made. Richard M. Schwend, MD, FAAP, Chairperson
could increase the likelihood of J. Eric Gordon, MD, FAAP
a fracture, the child may also have Norman Y. Otsuka, MD, FAAP
been abused because children with LEAD AUTHOR Ellen M. Raney, MD, FAAP
disabilities and other special health Emalee G. Flaherty, MD, FAAP Brian A. Shaw, MD, FAAP
Brian G. Smith, MD, FAAP
care needs are at increased risk of
Lawrence Wells, MD, FAAP
child abuse.83,84 Physicians should Paul W. Esposito, MD, USBJD Liaison
keep an open mind to the possibility of NEGLECT, 20122013
abuse and remember that child Cindy W. Christian, MD, FAAP, Chairperson STAFF
abuse occurs in all socioeconomic James E. Crawford-Jakubiak, MD, FAAP Niccole Alexander, MPP

PEDIATRICS Volume 133, Number 2, February 2014 e485

Downloaded from pediatrics.aappublications.org by guest on February 2, 2014
1. Loder RT, Feinberg JR. Orthopaedic inju- intentional injuries. Am J Dis Child. 1993; 27. Spevak MR, Kleinman PK, Belanger PL,
ries in children with nonaccidental 147(1):8792 Primack C, Richmond JM. Cardiopulmo-
trauma: demographics and incidence 14. Leventhal JM, Larson IA, Abdoo D, et al. nary resuscitation and rib fractures in
from the 2000 kids inpatient database Are abusive fractures in young children infants. A postmortem radiologic-
[published correction appears in J becoming less common? Changes over 24 pathologic study. JAMA. 1994;272(8):617
Pediatr Orthop. 2008;28(6):699]. J Pediatr years. Child Abuse Negl. 2007;31(3):311 618
Orthop. 2007;27(4):421426 322 28. Matshes EW, Lew EO. Two-handed cardio-
2. Jenny C, Hymel KP, Ritzen A, Reinert SE, 15. Kemp AM, Dunstan F, Harrison S, et al. pulmonary resuscitation can cause rib
Hay TC. Analysis of missed cases of abu- Patterns of skeletal fractures in child fractures in infants. Am J Forensic Med
sive head trauma [see comment; pub- abuse: systematic review [review]. BMJ. Pathol. 2010;31(4):303307
lished correction appears in JAMA. 1999; 2008;337:a1518 29. Kleinman PK, Perez-Rossello JM, Newton
282(1):29]. JAMA. 1999;281(7):621626 16. Hui C, Joughin E, Goldstein S, et al. Fem- AW, Feldman HA, Kleinman PL. Prevalence
3. Ravichandiran N, Schuh S, Bejuk M, et al. oral fractures in children younger than of the classic metaphyseal lesion in
Delayed identication of pediatric abuse- three years: the role of nonaccidental in- infants at low versus high risk for abuse.
related fractures. Pediatrics. 2010;125(1): jury. J Pediatr Orthop. 2008;28(3):297302 AJR Am J Roentgenol. 2011;197(4):1005
6066 17. Pandya NK, Baldwin K, Wolfgruber H, 1008
4. Skellern C, Donald T. Suspicious childhood Christian CW, Drummond DS, Hosalkar HS. 30. Kleinman PK, Marks SC, Jr, Richmond JM,
injury: formulation of forensic opinion. J Child abuse and orthopaedic injury pat- Blackbourne BD. Inicted skeletal injury:
Paediatr Child Health. 2011;47(11):771 terns: analysis at a level I pediatric a postmortem radiologic-histopathologic
775 trauma center. J Pediatr Orthop. 2009;29 study in 31 infants. AJR Am J Roent-
5. ONeill JA, Jr, Meacham WF, Grifn JP, (6):618625 genol. 1995;165(3):647650
Sawyers JL. Patterns of injury in the 18. Coffey C, Haley K, Hayes J, Groner JI. The 31. Kleinman PK. Problems in the diagnosis of
battered child syndrome. J Trauma. 1973; risk of child abuse in infants and toddlers metaphyseal fractures. Pediatr Radiol.
13(4):332339 with lower extremity injuries. J Pediatr 2008;38(suppl 3):S388S394
6. Gallagher SS, Finison K, Guyer B, Good- Surg. 2005;40(1):120123
32. OConnell A, Donoghue VB. Can classic
enough S. The incidence of injuries 19. Kleinman PK. The spectrum of non- metaphyseal lesions follow un-
among 87,000 Massachusetts children accidental injuries (child abuse) and its complicated caesarean section? Pediatr
and adolescents: results of the 198081 imitators. In: Hodler J, Zollikofer CL, Radiol. 2007;37(5):488491
Statewide Childhood Injury Prevention Schulthess GK, eds. Musculoskeletal Dis-
33. Grayev AM, Boal DK, Wallach DM, Segal LS.
Program Surveillance System. Am J Pub- eases 20092012. Milan, Italy: Springer
Metaphyseal fractures mimicking abuse
lic Health. 1984;74(12):13401347 Italia; 2009:227233
during treatment for clubfoot [see com-
7. Spady DW, Saunders DL, Schopocher DP, 20. Clarke NMP, Shelton FRM, Taylor CC, Khan
ment]. Pediatr Radiol. 2001;31(8):559563
Svenson LW. Patterns of injury in children: T, Needhirajan S. The incidence of frac-
a population-based approach. Pediatrics. 34. King J, Diefendorf D, Apthorp J, Negrete
tures in children under the age of 24
2004;113(3 pt 1):522529 VF, Carlson M. Analysis of 429 fractures in
monthsin relation to non-accidental
injury. Injury. 201243(6):762765 189 battered children. J Pediatr Orthop.
8. Rennie L, Court-Brown CM, Mok JYQ,
Beattie TF. The epidemiology of fractures 21. Kleinman PK. Diagnostic Imaging of Child
in children. Injury. 2007;38(8):913922 Abuse. 2nd ed. St. Louis, MO: Mosby; 1998 35. Pierce MC, Bertocci G. Injury bio-
mechanics and child abuse. Annu Rev
9. Leventhal JM, Martin KD, Asnes AG. In- 22. Barsness KA, Cha E-S, Bensard DD, et al.
Biomed Eng. 2008;10:85106
cidence of fractures attributable to abuse The positive predictive value of rib frac-
in young hospitalized children: results tures as an indicator of nonaccidental 36. Pierce MC, Bertocci GE, Vogeley E, Moreland
from analysis of a United States database. trauma in children [see comment]. J MS. Evaluating long bone fractures in chil-
Pediatrics. 2008;122(3):599604 Trauma. 2003;54(6):11071110 dren: a biomechanical approach with il-
10. Schwend RM, Werth C, Johnston A. Femur 23. Bulloch B, Schubert CJ, Brophy PD, Johnson lustrative cases. Child Abuse Negl. 2004;28
shaft fractures in toddlers and young N, Reed MH, Shapiro RA. Cause and clinical (5):505524
children: rarely from child abuse. J characteristics of rib fractures in infants. 37. Pierce MC, Bertocci G. Fractures resulting
Pediatr Orthop. 2000;20(4):475481 Pediatrics. 2000;105(4). Available at: www. from inicted trauma: assessing injury
11. Hennrikus WL, Shaw BA, Gerardi JA. Inju- pediatrics.org/cgi/content/full/105/4/E48 and history compatibility. Clin Pediatr
ries when children reportedly fall from 24. Kleinman PK, Schlesinger AE. Mechanical Emerg Med. 2006;7(3):143148
a bed or couch. Clin Orthop Relat Res. factors associated with posterior rib 38. Rex C, Kay PR. Features of femoral frac-
2003; (407):148151 fractures: laboratory and case studies. tures in nonaccidental injury. J Pediatr
12. Worlock P, Stower M, Barbor P. Patterns of Pediatr Radiol. 1997;27(1):8791 Orthop. 2000;20(3):411413
fractures in accidental and non-accidental 25. Bixby SD, Abo A, Kleinman PK. High-impact 39. Haney SB, Starling SP, Heisler KW, Okwara
injury in children: a comparative study. Br trauma causing multiple posteromedial L. Characteristics of falls and risk of in-
Med J (Clin Res Ed). 1986;293(6539):100 rib fractures in a child. Pediatr Emerg jury in children younger than 2 years.
102 Care. 2011;27(3):218219 Pediatr Emerg Care. 2010;26(12):914918
13. Leventhal JM, Thomas SA, Roseneld NS, 26. Feldman KW, Brewer DK. Child abuse, 40. Nimityongskul P, Anderson LD. The likeli-
Markowitz RI. Fractures in young children. cardiopulmonary resuscitation, and rib hood of injuries when children fall out of
Distinguishing child abuse from un- fractures. Pediatrics. 1984;73(3):339342 bed. J Pediatr Orthop. 1987;7(2):184186


Downloaded from pediatrics.aappublications.org by guest on February 2, 2014

41. Lyons TJ, Oates RK. Falling out of bed: 55. Wallis GA, Starman BJ, Zinn AB, Byers PH. 70. Gordon CM, Feldman HA, Sinclair L, et al.
a relatively benign occurrence. Pediatrics. Variable expression of osteogenesis Prevalence of vitamin D deciency among
1993;92(1):125127 imperfecta in a nuclear family is healthy infants and toddlers. Arch Pediatr
42. Hansoti B, Beattie T. Can the height of fall explained by somatic mosaicism for a le- Adolesc Med. 2008;162(6):505512
predict long bone fracture in children thal point mutation in the alpha 1(I) gene 71. Greenspan A. Orthopedic Imaging: A
under 24 months? Eur J Emerg Med. 2005; (COL1A1) of type I collagen in a parent. Am
Practical Approach. 4th ed. Philadelphia,
12(6):285286 J Hum Genet. 1990;46(6):10341040
PA: Lippincott Williams & Wilkins; 2004
43. Thomas SA, Roseneld NS, Leventhal JM, 56. Barsh GS, Byers PH. Reduced secretion of
72. Perez-Rossello JM, Feldman HA, Kleinman
Markowitz RI. Long-bone fractures in structurally abnormal type I procollagen
PK, et al. Rachitic changes, deminer-
young children: distinguishing accidental in a form of osteogenesis imperfecta.
Proc Natl Acad Sci USA. 1981;78(8):5142 alization, and fracture risk in healthy
injuries from child abuse. Pediatrics. infants and toddlers with vitamin D
1991;88(3):471476 5146
57. Sprigg A. Temporary brittle bone disease deciency. Radiology. 2012;262(1):234
44. Loder RT, ODonnell PW, Feinberg JR. Epi- 241
demiology and mechanisms of femur versus suspected non-accidental skeletal
injury. Arch Dis Child. 2011;96(5):411413 73. Slovis TL, Chapman S. Vitamin D in-
fractures in children. J Pediatr Orthop.
58. Greeley CS, Donaruma-Kwoh M, Vettimattam sufciency/deciencya conundrum.
M, Lobo C, Williard C, Mazur L. Fractures at Pediatr Radiol. 2008;38(11):1153
45. Baldwin K, Pandya NK, Wolfgruber H,
diagnosis in infants and children with 74. Slovis TL, Chapman S. Evaluating the data
Drummond DS, Hosalkar HS. Femur frac-
osteogenesis imperfecta. J Pediatr Orthop. concerning vitamin D insufciency/
tures in the pediatric population: abuse
2013;33(1):3236 doi:10.1097/BPO.1090- deciency and child abuse. Pediatr
or accidental trauma? Clin Orthop Relat
b1013e318279c318255d Radiol. 2008;38(11):12211224
Res. 2011;469(3):798804
59. Singh Kocher M, Dichtel L. Osteogenesis 75. Jenny C. Rickets or abuse? Pediatr Radiol.
46. Pierce MC, Bertocci GE, Janosky JE, et al.
imperfecta misdiagnosed as child abuse. 2008;38(11):12191220
Femur fractures resulting from stair falls
J Pediatr Orthop B. 2011;20(6):440443 76. Schilling S, Wood JN, Levine MA, Langdon
among children: an injury plausibility
60. Gahagan S, Rimsza ME. Child abuse or D, Christian CW. Vitamin D status in
model. Pediatrics. 2005;115(6):17121722
osteogenesis imperfecta: how can we tell? abused and nonabused children younger
47. Haney SB, Boos SC, Kutz TJ, Starling SP. Pediatrics. 1991;88(5):987992
Transverse fracture of the distal femoral than 2 years old with fractures. Pediat-
61. Ablin DS, Sane SM. Non-accidental in- rics. 2011;127(5):835841
metadiaphysis: a plausible accidental
jury: confusion with temporary brittle
mechanism. Pediatr Emerg Care. 2009;25 77. Chapman T, Sugar N, Done S, Marasigan J,
bone disease and mild osteogenesis
(12):841844 Wambold N, Feldman K. Fractures in
imperfecta. Pediatr Radiol. 1997;27(2):
48. Grant P, Mata MB, Tidwell M. Femur frac- 111113 infants and toddlers with rickets. Pediatr
ture in infants: a possible accidental eti- Radiol. 2010;40(7):11841189
62. Knight DJ, Bennet GC. Nonaccidental injury
ology. Pediatrics. 2001;108(4):10091011 in osteogenesis imperfecta: a case report. 78. Perez-Rossello JM, McDonald AG, Rosenberg
49. Pandya NK, Baldwin KD, Wolfgruber H, J Pediatr Orthop. 1990;10(4):542544 AE, Ivey SL, Richmond JM, Kleinman PK.
Drummond DS, Hosalkar HS. Humerus 63. Jenny C; Committee on Child Abuse and Prevalence of rachitic changes in deceased
fractures in the pediatric population: an Neglect. Evaluating infants and young infants: a radiologic and pathologic study.
algorithm to identify abuse. J Pediatr children with multiple fractures. Pediat- Pediatr Radiol. 2011;41(suppl 1):S57
Orthop B. 2010;19(6):535541 rics. 2006;118(3):12991303 79. Ogden JA. Pediatric osteomyelitis and
50. Strait RT, Siegel RM, Shapiro RA. Humeral 64. Backstrm MC, Kuusela A-L, Mki R. Met- septic arthritis: the pathology of neonatal
fractures without obvious etiologies in abolic bone disease of prematurity. Ann disease. Yale J Biol Med. 1979;52(5):423
children less than 3 years of age: when is Med. 1996;28(4):275282 448
it abuse? Pediatrics. 1995;96(4 pt 1):667 65. Naylor KE, Eastell R, Shattuck KE, Alfrey AC, 80. Whedon GD. Disuse osteoporosis: physio-
671 Klein GL. Bone turnover in preterm logical aspects. Calcif Tissue Int. 1984;36
51. Hymel KP, Jenny C. Abusive spiral frac- infants. Pediatr Res. 1999;45(3):363366 (suppl 1):S146S150
tures of the humerus: a videotaped ex- 66. Harrison CM, Johnson K, McKechnie E. 81. Presedo A, Dabney KW, Miller F. Fractures
ception. Arch Pediatr Adolesc Med. 1996; Osteopenia of prematurity: a national in patients with cerebral palsy. J Pediatr
150(2):226227 survey and review of practice. Acta Pae- Orthop. 2007;27(2):147153
52. Laskey AL, Stump TE, Hicks RA, Smith JL. diatr. 2008;97(4):407413
82. Westcott H. The abuse of disabled chil-
Yield of skeletal surveys in children 18 67. Amir J, Katz K, Grunebaum M, Yosipovich
months of age presenting with isolated dren: a review of the literature. Child Care
Z, Wielunsky E, Reisner SH. Fractures in
skull fractures. J Pediatr. 2013;162(1):86 Health Dev. 1991;17(4):243258
premature infants. J Pediatr Orthop. 1988;
89 8(1):4144 83. Sullivan PM, Knutson JF. The association
53. Wood JN, Christian CW, Adams CM, Rubin between child maltreatment and dis-
68. Bugental DB, Happaney K. Predicting in-
DM. Skeletal surveys in infants with iso- fant maltreatment in low-income families: abilities in a hospital-based epidemiolog-
lated skull fractures. Pediatrics. 2009;123 the interactive effects of maternal attri- ical study. Child Abuse Negl. 1998;22(4):
(2). Available at: www.pediatrics.org/cgi/ butions and child status at birth. Dev 271288
content/full/123/2/e247e252 Psychol. 2004;40(2):234243 84. Sullivan PM, Knutson JF. Maltreatment
54. Byers PH, Steiner RD. Osteogenesis 69. Keller KA, Barnes PD. Rickets vs. abuse: and disabilities: a population-based epi-
imperfecta. Annu Rev Med. 1992;43:269 a national and international epidemic. demiological study. Child Abuse Negl.
282 Pediatr Radiol. 2008;38(11):12101216 2000;24(10):12571273

PEDIATRICS Volume 133, Number 2, February 2014 e487

Downloaded from pediatrics.aappublications.org by guest on February 2, 2014
85. Olmedo JM, Yiannias JA, Windgassen EB, clinical ndings. Pediatrics. 1986;78(5): in suspected child abuse. Pediatr Radiol.
Gornet MK. Scurvy: a disease almost for- 803807 2013;43(5):575581
gotten. Int J Dermatol. 2006;45(8):909913 101. Taitz J, Moran K, OMeara M. Long bone 114. Karmazyn B, Duhn RD, Jennings SG, et al.
86. Larralde M, Santos Muoz A, Boggio P, Di fractures in children under 3 years of age: Long bone fracture detection in suspected
Gruccio V, Weis I, Schygiel A. Scurvy in is abuse being missed in emergency de- child abuse: contribution of lateral views
a 10-month-old boy. Int J Dermatol. 2007; partment presentations? J Paediatr Child [published online ahead of print October
46(2):194198 Health. 2004;40(4):170174 6, 2011]. Pediatr Radiol. doi: 10.1007/
87. Nishio H, Matsui K, Tsuji H, Tamura A, 102. Farrell C, Rubin DM, Downes K, Dormans J, s00247-011-2248-3
Suzuki K. Immunohistochemical study of Christian CW. Symptoms and time to 115. van Rijn RR. How should we image skeletal
tyrosine phosphorylation signaling in the medical care in children with accidental injuries in child abuse? Pediatr Radiol.
involuted thymus. Forensic Sci Int. 2000; extremity fractures. Pediatrics. 2012;129 2009;39(suppl 2):S226S229
110(3):189198 (1). Available at: www.pediatrics.org/cgi/ 116. Kleinman PK, Nimkin K, Spevak MR, et al.
88. Shaw JC. Copper deciency and non- content/full/129/1/ e128133 Follow-up skeletal surveys in suspected
accidental injury. Arch Dis Child. 1988;63 103. Peters ML, Starling SP, Barnes-Eley ML, child abuse. AJR Am J Roentgenol. 1996;
(4):448455 Heisler KW. The presence of bruising as- 167(4):893896
89. Marquardt ML, Done SL, Sandrock M, sociated with fractures. Arch Pediatr 117. Bennett BL, Chua MS, Care M, Kachelmeyer
Berdon WE, Feldman KW. Copper de- Adolesc Med. 2008;162(9):877881 A, Mahabee-Gittens M. Retrospective review
ciency presenting as metabolic bone 104. Valvano TJ, Binns HJ, Flaherty EG, Leon- to determine the utility of follow-up skele-
disease in extremely low birth weight, hardt DE. Does bruising help determine tal surveys in child abuse evaluations
short-gut infants. Pediatrics. 2012;130(3). which fractures are caused by abuse? when the initial skeletal survey is normal.
Available at: www.pediatrics.org/cgi/con- Child Maltreat. 2009;14(4):376381 BMC Res Notes. 2011;4(1):354
tent/full/130/3/e695e698 105. Sugar NF, Taylor JA, Feldman KW; Puget 118. Harlan SR, Nixon GW, Campbell KA, Hansen
90. Tmer Z, Mller LB. Menkes disease. Eur J Sound Pediatric Research Network. K, Prince JS. Follow-up skeletal surveys
Hum Genet. 2010;18(5):511518 Bruises in infants and toddlers: those who for nonaccidental trauma: can a more
dont cruise rarely bruise. Arch Pediatr limited survey be performed? Pediatr
91. Bacopoulou F, Henderson L, Philip SG.
Adolesc Med. 1999;153(4):399403 Radiol. 2009;39(9):962968
Menkes disease mimicking non-accidental
injury [published correction appears in 106. Pierce MC, Kaczor K, Aldridge S, OFlynn J, 119. Mandelstam SA, Cook D, Fitzgerald M,
Arch Dis Child. 2009;94(1):77]. Arch Dis Lorenz DJ. Bruising characteristics dis- Ditcheld MR. Complementary use of ra-
criminating physical child abuse from diological skeletal survey and bone scin-
Child. 2006;91(11):919
accidental trauma. Pediatrics. 2010;125 tigraphy in detection of bony injuries in
92. Paterson CR, Burns J, McAllion SJ. Osteo-
(1):6774 suspected child abuse. Arch Dis Child.
genesis imperfecta: the distinction from
107. Shapiro JR, Sponsellor PD. Osteogenesis 2003;88(5):387390, discussion 387390
child abuse and the recognition of a vari-
imperfecta: questions and answers. Curr 120. Drubach LA, Johnston PR, Newton AW,
ant form [see comment]. Am J Med Genet.
Opin Pediatr. 2009;21(6):709716 Perez-Rossello JM, Grant FD, Kleinman PK.
108. Marlowe A, Pepin MG, Byers PH. Testing Skeletal trauma in child abuse: detection
93. Chapman S, Hall CM. Non-accidental injury
for osteogenesis imperfecta in cases of with 18F-NaF PET. Radiology. 2010;255(1):
or brittle bones. Pediatr Radiol. 1997;27
suspected non-accidental injury. J Med 173181
Genet. 2002;39(6):382386 121. Rubin DM, Christian CW, Bilaniuk LT,
94. Paterson CR. Temporary brittle bone dis-
109. American Academy of Pediatrics; Section Zazyczny KA, Durbin DR. Occult head injury
ease: fractures in medical care. Acta
on Radiology; American Academy of Pedi- in high-risk abused children. Pediatrics.
Paediatr. 2009;98(12):19351938
atrics. Diagnostic imaging of child abuse. 2003;111(6 pt 1):13821386
95. Paterson CR. Temporary brittle bone dis- Pediatrics. 2009;123(5):14301435 122. Wootton-Gorges SL, Stein-Wexler R, Walton
ease: the current position. Arch Dis Child.
110. Duffy SO, Squires J, Fromkin JB, Berger RP. JW, Rosas AJ, Coulter KP, Rogers KK.
Use of skeletal surveys to evaluate for Comparison of computed tomography and
96. Miller ME. The lesson of temporary brittle physical abuse: analysis of 703 consecu- chest radiography in the detection of rib
bone disease: all bones are not created tive skeletal surveys. Pediatrics. 2011;127 fractures in abused infants. Child Abuse
equal. Bone. 2003;33(4):466474 (1). Available at: www.pediatrics.org/cgi/ Negl. 2008;32(6):659663
97. Miller ME. Temporary brittle bone disease: content/full/127/1/e47e52 123. Brody AS, Frush DP, Huda W, Brent RL;
a true entity? Semin Perinatol. 1999;23(2): 111. American College of Radiology. ACR Prac- American Academy of Pediatrics Section
174182 tice Guidelines for Skeletal Surveys in on Radiology. Radiation risk to children
98. Miller ME, Hangartner TN. Temporary Children. 2011. Available at: www.acr.org/ from computed tomography. Pediatrics.
brittle bone disease: association with de- /media/ACR/Documents/PGTS/guidelines/ 2007;120(3):677682
creased fetal movement and osteopenia. Skeletal_Surveys.pdf. Accessed May 5, 2013 124. Stranzinger E, Kellenberger CJ, Braunschweig
Calcif Tissue Int. 1999;64(2):137143 112. Ingram JD, Connell J, Hay TC, Strain JD, S, Hopper R, Huisman TAGM. Whole-body
99. Palacios J, Rodriguez JI. Extrinsic fetal MacKenzie T. Oblique radiographs of the STIR MR imaging in suspected child abuse:
akinesia and skeletal development: chest in nonaccidental trauma. Emerg an alternative to skeletal survey radiog-
a study in oligohydramnios sequence. Radiol. 2000;7(1):4246 raphy? Eur J Radiol Extra. 2007;63(1):43
Teratology. 1990;42(1):15 113. Prabhu SP, Newton AW, Perez-Rossello 47
100. Rivara FP, Parish RA, Mueller BA. Extremity JM, Kleinman PK. Three-dimensional 125. Perez-Rossello JM, Connolly SA, Newton
injuries in children: predictive value of skull models as a problem-solving tool AW, Zou KH, Kleinman PK. Whole-body MRI


Downloaded from pediatrics.aappublications.org by guest on February 2, 2014

in suspected infant abuse. AJR Am J Clin Endocrinol Metab. 2011;96(10):3160 tigators. Prevalence of abusive injuries in
Roentgenol. 2010;195(3):744750 3169 siblings and household contacts of phys-
126. Bachrach LK, Sills IN; Section on Endocri- 129. Gordon CM, Baim S, Bianchi M-L, et al; ically abused children. Pediatrics. 2012;
nology. Clinical reportbone densitome- International Society for Clinical Densi- 130(2):193201
try in children and adolescents. tometry. Special report on the 2007 Pedi- 132. Hamilton-Giachritsis CE, Browne KD. A
Pediatrics. 2011;127(1):189194 atric Position Development Conference of retrospective study of risk to siblings in
127. Khoury DJ, Szalay EA. Bone mineral den- the International Society for Clinical Den- abusing families. J Fam Psychol. 2005;19
sity correlation with fractures in non- sitometry. South Med J. 2008;101(7):740 (4):619624
ambulatory pediatric patients. J Pediatr 743 133. Asnes AG, Leventhal JM. Managing child
Orthop. 2007;27(5):562566 130. Henderson RC, Lark RK, Newman JE, et al. abuse: general principles. Pediatr Rev.
128. Zemel BS, Kalkwarf HJ, Gilsanz V, et al. Pediatric reference data for dual x-ray 2010;31(2):4755
Revised reference curves for bone min- absorptiometric measures of normal 134. Banaszkiewicz PA, Scotland TR, Myerscough
eral content and areal bone mineral bone density in the distal femur. AJR Am J EJ. Fractures in children younger than age
density according to age and sex for black Roentgenol. 2002;178(2):439443 1 year: importance of collaboration with
and non-black children: results of the 131. Lindberg DM, Shapiro RA, Laskey AL, Pallin child protection services. J Pediatr Orthop.
bone mineral density in childhood study. J DJ, Blood EA, Berger RP; ExSTRA Inves- 2002;22(6):740744

PEDIATRICS Volume 133, Number 2, February 2014 e489

Downloaded from pediatrics.aappublications.org by guest on February 2, 2014
Evaluating Children With Fractures for Child Physical Abuse
Emalee G. Flaherty, Jeannette M. Perez-Rossello, Michael A. Levine, William L.
Pediatrics 2014;133;e477; originally published online January 27, 2014;
DOI: 10.1542/peds.2013-3793
Updated Information & including high resolution figures, can be found at:
Services http://pediatrics.aappublications.org/content/133/2/e477.full.h
References This article cites 125 articles, 35 of which can be accessed
free at:
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
Reprints Information about ordering reprints can be found online:

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
Grove Village, Illinois, 60007. Copyright 2014 by the American Academy of Pediatrics. All
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org by guest on February 2, 2014