Epidemiology of Fractures in Children

Yes, Results are Definitely Better for Many Injuries,
such as Supracondylar Humeral Fractures. The superiority of operative
treatment of supracondylar fractures of the distal humerus was clearly
demonstrated in a report published in 1988 from Toronto, Canada, in
which treatment in traction, treatment with a cast alone, and treatment
with percutaneous pin fixation were compared.⁹²
The worst results were in patients treated with only a cast. The best
results were achieved in those stabilized with percutaneous pin
fixation. The universal acceptance of percutaneous pin fixation of
these fractures is evidence of the superiority of operative management.

Cox and Clarke, in evaluating the fracture management in
their hospital in Southampton, England, found a high incidence of
secondary hospital treatment for fractures initially managed
nonoperatively.²⁸ There was a 12%
readmission rate to correct late displacement of fractures of the
radius and distal humerus. In addition, 24% of their internal fixation
procedures were to salvage unacceptable results of nonoperative
management. They concluded that more selective initial operative
intervention in radial and distal humeral fractures could decrease the
incidence of costly readmissions to the hospital.

Maybe, Depending on What You Call Results. A 7-year-old
with a midshaft fracture of the femur may have had the same excellent
bony alignment and healing when treated with 6 weeks of skeletal
traction as when treated with IM fixation. However, quality of life
during treatment, burden of care on the family, and costs are markedly
different in these two scenarios.

In some cases, operative fixation has created a new set
of iatrogenic problems that result in less favorable outcomes for some
children. Some of the specific problems that have occurred over the
years are: (i) ulnar nerve injury with medial pin fixation of
supracondylar fractures,⁶⁷ (ii) high refracture rate with external fixation of femoral shaft fractures,⁹⁴ and (iii) osteonecrosis of the femoral head following the use of interlocking IM nails inserted through the piriformis fossa.⁹,⁷⁷

With emphasis on operative management, the fact that
most children’s fractures can be managed by nonoperative techniques has
become obscured. As a result, many recent orthopaedic trainees are less
exposed to and less comfortable with nonoperative technical skills.

In fact, several articles have demonstrated excellent
results of treating children’s fractures by focusing on improvements in
nonoperative methods, “pleading for conservatism.”⁴⁷ Chess et al.²³ showed that when properly applied, a well-molded short-arm

cast provides just as good a result as a long-arm cast in treating
displaced fractures of the distal radial metaphysis. These authors
believed the key to success in using a short-arm cast is in a careful
molding of the cast at the fracture site so there is the proper cast
index of 0.7 or less. Walker and Rang challenged traditional thinking
by demonstrating that unstable fractures of the radius and ulna could
be treated with a lower frequency of remanipulation if immobilized in
elbow extension rather than flexion.¹²⁷

It is important to remember that most children’s fractures are still treated by nonoperative methods.

The incidence of pediatric fracture varies in different
cultural settings. For instance, Cheng and Shen studied children in
Hong Kong who lived in confined high-rise apartments.²²
Their risk of exposure to injury differed from the study by Reed of
children living in the rural environment of Winnipeg, Canada.⁹⁷ Two separate reviews by Laffoy⁵⁵ and Westfelt⁸⁶
found that children in a poor social environment (as defined by a lower
social class or by dependence on public assistance) had more frequent
accidents than more affluent children. In England, children from
single-parent families were found to have higher accident and infection
rates than children from two parent families.³⁶

Two additional studies in the United Kingdom looked at
the relationship of affluence to the incidence of fractures in
children. Lyons et al.⁶⁸ found no
difference in the fracture rates of children in affluent population
groups compared to those of children in nonaffluent families. On the
other hand, Stark et al.¹¹⁹ in
Scotland found that the fracture rates in children from nonaffluent
social groups was significantly higher than those in affluent families.

The climate may be a strong factor as well. Children in
colder climates, with ice and snow, are exposed to risks different from
those of children living in warmer climates. The exposure time to
outdoor activities may be greater for children who live in warmer
climates. For example, the incidence of chronic overuse elbow injuries
in young baseball players (Little League elbow) is far greater in the
southern United States than in the northern part of the country.

Pediatric trauma should be viewed as a disease where
there are direct and predictable relationships between exposure and
incidence.

Several sources of administrative, national, and
regional data have recently become available providing significantly
improved investigation into various areas within pediatric trauma. The
Healthcare Cost and Utilization Project (HCUP) is a family of databases
including the State Inpatient Databases (SID), the Nationwide Inpatient
Sample (NIS), and the Kids’ Inpatient Database (KID). While
administrative data may lack clinical detail for certain purposes,
these datasets provide a comprehensive overview of healthcare
utilization in the United States and are available without purchase
(http://www.ahrq.gov/data/hcup/hcupnet.htm).¹²⁰
The KID database has been increasingly used to examine the incidence of
pediatric trauma as well as practice patterns in pediatric trauma. Data
for KIDS are collected and published every 3 years, with data currently
available for 1997, 2000, 2003, and 2006. KIDS is “nationally
representatative,” meaning that the database contains a large but
incomplete sample of the hospital discharge records (3.1 million in
2006), which are then statisticaly weighted upward to reflect the
complete population of pediatric discharges (7.6 million in 2006).
Several other databases including the National Electronic Injury
Surveillance System (http://www.cpsc.gov/library/neiss.html) have also
been useful in providing information about the epidemiology of
pediatric trauma.

Currently available data sources provide scant clinical
detail, limiting broader utility as a source of health outcomes data in
the field. Constructed in an attempt to fill such a role, the National
Pediatric Trauma Registry (NPTR) is a multi-institutional database
designed to provide a snapshot of physiological and clinical
information. The NPTR was functional for about 15 years and provided a
source of important data in the realm of pediatric trauma.¹²²
The NPTR is currently being redesigned into an even more powerful
database that will be called the National Trauma Registry for Children,
which should serve as a powerful reference for contributors to future
editions of this book.

Early reviews primarily developed a knowledge base of
fracture healing in children. In 1941, Beekman and Sullivan published
an extensive review of the incidence of children’s fractures.¹¹
Their pioneering work—still quoted today—included a study of 2094 long
bone fractures seen over a 10-year period at Bellevue Hospital in New
York City. The major purpose of their study was to develop basic
principles for treating children’s fractures.

In 1954, two reports, one by Hanlon and Estes⁴¹ and the other by Lichtenberg,⁶²
confirmed the findings of the previous studies with regard to the
general incidence of children’s long bone fractures and their ability
to heal and readily remodel. These initial reviews were mainly
statistical analyses and did not delve deeply into the true
epidemiology of children’s fractures. In 1965, Wong explored the effect
of cultural factors on the incidence of fractures by comparing Indian,
Malay, and Swedish children.¹³³ In the 1970s, two other studies, one by Iqbal⁴⁴ and another by Reed,⁹⁷ added more statistics regarding the incidence of the various long bone fractures.

Landin’s 1983 report on 8682 fractures remains a landmark on this subject.⁵⁸
He reviewed the data on all fractures in children that occurred in
Malmo, Sweden, over 30 years and examined the factors affecting the
incidence of children’s fractures. By studying two populations, 30
years apart, he determined that fracture patterns were changing and
suggested reasons for such changes. His initial goal was to establish
data for preventive programs, so he focused on fractures that produced
clean, concise, concrete data.

In 1997, Landin updated his work, re-emphasizing the statistics from his previous publication.⁵⁷
He suggested that the twofold increase in fracture rate during the 30
years from 1950 to 1979 in Malmo was due mainly to an increased
participation in sports. In 1999, in cooperation with Tiderius and
Duppe, Landin¹²³ studied the
incidence in the same age group again in Malmo and found that the rate
had actually declined by 9% in 1993 and 1994. The only exception was an
increase of distal forearm fractures in girls, which he attributed to
their increased participation in sporting events.

Cheng and Shen,⁶⁶ in
their 1993 study from Hong Kong, also set out to define children’s
fractures by separating the incidences into age groups. They tried to
gather epidemiologic data on which to build preventive programs. In
1999, this study was expanded to include almost 6500 fractures in
children 16 and younger over a 10-year period.²¹
The fracture patterns changed little over those 10 years. What did
change was the increased frequency of closed reduction and percutaneous
pin fixation of fractures, with a corresponding decrease in open
reductions. There also was a marked decrease in the hospital stay of
their patients.

More recently, using the HCUP’s KIDS dataset, Galano et al.⁴⁰
examined the face of pediatric inpatient trauma in 1997. They estimated
that roughly 84,000 children were admitted for fracture care which
resulted in about 1 billion dollars in hospital charges. Of some
interest, more than 70% of children were treated at non-children’s
hospitals.

In Landin’s series from Malmo, Sweden, the chance of a
child sustaining a fracture during childhood (birth to age 16) was 42%
for boys and 27% for girls.⁵⁸ When
considered on an annual basis, 2.1% of all the children (2.6% for boys;
1.7% for girls) sustained at least one fracture each year. These
figures were for all fracture types and included those treated on an
inpatient basis and an outpatient basis. The overall chance of fracture
per year was 1.6% for both girls and boys in a study from England of
both outpatients and inpatients by Worlock and Stower.¹³⁴
The chance of a child sustaining a fracture severe enough to require
inpatient treatment during the first 16 years of life is 6.8%.²²
Thus, on an annual basis, 0.43% of the children in an average community
will be admitted for a fracture-related problem during the year.

In a series of 23,915 patients seen at four major hospitals for injury-related complaints, 4265 (17.8%) had fractures.¹⁷,⁴¹,⁷⁵,⁸⁶ Thus, close to 20% of the patients who present to hospitals with injuries have a fracture.

It is interesting to note that, in a follow-up study by Tiderius, Landin, and Duppe¹²³ in the years 1993 and 1994, 13 years after the termination of the original 30-year study by Landin,⁵⁸
there was an almost 10% decrease in the incidence of fractures in the
0- to 16-year age group. They attributed this to less physical activity
on the part of modern-day children coupled with better protective
sports equipment and increased traffic safety (e.g., stronger cars and
use of auto restraint systems). The overall incidence of children’s
fractures is summarized in Table 1-1.

Fractures Show a Linear Increase with Age. Starting with
birth and extending to age 12, all the major series that segregated
patients by age have demonstrated a linear increase in the annual
incidence of fractures with age (Fig. 1-1).¹⁶,²¹,²²,⁴⁴,⁵⁸,¹³⁴

Although there is a high incidence of injuries in children ages 1 to 2, the incidence of fractures is low.⁵⁵
Most injuries in children of this age are nonorthopaedic entities such
as head injuries, lacerations, and abrasions. In fact, the incidence of
lacerations in both sexes peaks at this age.⁹⁹

In 1962, Kempe et al.⁴⁹
called attention to the frequency of fractures and other injuries in
young children that were due to nonaccidental trauma. They termed these
injuries part of the

battered child syndrome. Arkbania et al.²
later defined the specific fracture patterns seen in victims of child
abuse. The high rate of fractures from nonaccidental trauma has been
shown to extend to age 3.⁵²

Not all fractures in the first year of life can be
attributed to abuse. In a review of fractures occurring in the first
year of life, McClelland and Heiple found that fully 44% were from
documented accidental and nonabusive etiologies.⁷⁴
They also noted that 23% of these patients had generalized conditions
that predisposed them to fractures. Thus, although nonaccidental trauma
remains the leading cause of fractures during the first year of life,
other general and metabolic conditions may predispose children to
fractures from accidental causes.

Information from the 2000 KIDS database indicates that
about half of abused hospitalized children older than 3 years of age
have concomitant psychiatric or neurological conditions, reminding
caretakers to maintain vigilance in this at risk population.⁶⁵
For example, a nonambulatory child with cerebral palsy is expected to
have osteopenia and be at increasd risk for fracture. The orthopaedic
surgeon should not fall into the trap, however, of assuming that all
fractures in children with cerebral palsy are accidents, because
children with cerebral palsy also are at an increased risk of child
abuse.

Males Predominate in Late Age Groups. The male predominance of injury and fracture victims has been discussed (see Table 1-1; Figs. 1-1 and 1-2). For all age groups, the overall ratio of boys to girls who sustain a single fracture is 2.7:1.²² In girls, fracture incidence peaks just before adolescence and then decreases during adolescence.²²,⁵⁸,⁹⁷ In the 10-year study from Hong Kong by Chang et al.,²¹
the male incidence in the 12- to 16-year age group was 83%. The
incidence of fractures in girls steadily declined from their peak in
the birth to 3-year age group.

In some areas, there is little difference in the
incidence of fractures between boys and girls. For example, during the
first 2 years of life, the overall incidence of injuries and fractures
in both genders is nearly equal. During these first 2 years, the injury
rates for foreign body ingestion, poisons, and burns have no
significant gender differences. With activities in which there is a
male difference in participation, such as with sports equipment and
bicycles, there is a marked increase in the incidence of injuries in
boys.²¹,⁹⁹

Role of Behavior. The injury incidence may not be due to
the rate of exposure alone; behavior may be a major factor. For
example, one study found that the incidence of auto/pedestrian
childhood injuries peaks in both sexes at ages 5 to 8.¹⁰⁴
When the total number of street crossings per day was studied, both
sexes did so equally. Despite this equal exposure, boys had a higher
number of injuries. Thus, the difference in the rate between the sexes
begins to develop a male predominance when behaviors change. The
difference in the injury rate between the genders may change in the
future as more girls participate in activities with increased physical
risk.²¹,⁹⁹

In most series, the left upper extremity demonstrates a slight but significant predominance.¹⁴,²⁷,³⁰,³¹,³⁴,³⁹
The ratio of left to right overall averages 1.3:1. In some fractures,
however, especially those of supracondylar bones, lateral condyles, and
the distal radius, the incidence is far greater, increasing to as much
as 2.3:1 for the lateral condyle. In the lower extremity, the incidence
of injury on the right side is slightly increased.⁴¹,⁵⁸

The reasons for the predominance of the left upper extremity have been studied, but no definite answers have been found. Rohl¹⁰²
speculated that the right upper extremity is often being used actively
during the injury, so the left assumes the role of protection. In a
study examining the left-sided predominance in the upper extremity,
Mortensson and Thonell⁸⁰ questioned
patients and their parents on arrival to the emergency department about
which arm was used for protection and the position of the fractured
extremity at the time of the accident. They

found
two trends: regardless of handedness, the left arm was used more often
to break the fall, and when exposed to trauma, the left arm was more
likely to be fractured.

Summertime Increase. Fractures are more common during
the summer, when children are out of school and exposed to more
vigorous physical activities (Fig. 1-3). Five studies from the northern hemisphere have confirmed this summertime increase.²¹,²²,¹⁰²,¹²⁹,¹³⁴

Hours of Sunshine. The most consistent climatic factor appears to be the number of hours of sunshine. Masterson et al,⁷³
in a study from Ireland, found a strong positive correlation between
monthly sunshine hours and monthly fracture admissions. There was also
a weak negative correlation with monthly rainfall. Overall, the average
number of fractures in the summer was 2.5 times that in the winter. In
days with more sunshine hours than average, the average fracture
admission rate was 2.31 per day; on days with fewer sunshine hours than
average, the admission rate was 1.07 per day.

In Sweden, the incidence of fractures in the summer had
a bimodal pattern that seemed to be influenced by cultural traditions.
In two large series of both accidents and fractures in Sweden by
Westfelt⁸⁶ and Landin,⁵⁸
the researchers noticed increases in May and September and significant
decreases in June, July, and August. Both writers attributed this to
the fact that children in their region left the cities to spend the
summer in the countryside. Thus, the decrease in the overall fracture
rate was probably due to a decrease in the number of children at risk
remaining in the city.

Masterson et al.⁷³
speculated that because the rate of growth increases during the summer,
the number of physeal fractures should also increase, as the physes
would be weaker during this time. For example, the incidence of slipped
capital femoral epiphysis, which is related to physeal weakness,
increases during the summer.⁷
However, Landin, in his study of more than 8000 fractures of all types,
found the overall seasonal incidence of physeal injuries to be exactly
the same as nonphyseal injuries.⁵⁸

Younger Age Groups Unaffected. Thus, it appears that
climate, especially in areas where there are definite seasonal
variations, influences the incidence of fractures in all children,
especially in older children. However, in small children and infants,
whose activities are not seasonally dependent, there appears to be no
significant seasonal influence.

The time of day in which children are most active seems
to correlate with the peak time for fracture occurrence. In Sweden, the
incidence peaked between 2 and 3 PM.⁸⁶
In a well-documented study from Texas by Shank et al.,110 the hourly
incidence of fractures formed a well-defined bell curve peaking at
about 6 PM (Fig. 1-4).

Increase in Minor Trauma. Landin’s study is the only one
that has compared the changes over a significant time span: his data
were collected over 30 years.⁵⁸ He
classified the degree of trauma as slight, moderate, or severe. The
incidence of all trauma in both boys and girls increased significantly
over the 30-year study period, but the incidence of severe trauma
increased only slightly. The greatest increase was in the “slight”
category. Landin attributed the increase in this category to the
introduction of subsidized medical care. Because expense was not a
factor, parents were more inclined in the later years of the study to
seek medical attention for relatively minor complaints. Physicians,
likewise, were more inclined to order radiographs. Thus, many of the
minor injuries, such as torus fractures, which were often ignored in
the earlier years, were seen more often at medical facilities during
the later years.

Likewise, the overall incidence of fractures in Malmo, Sweden (the same city as Landin’s original study),⁵⁸ had decreased significantly (10%) in the more recent years.¹²³

The one fracture type that exhibited a true increase
over this period was that of the femoral shaft. This increase was
thought to be influenced by new types of play activities and increased
participation in sports.

Increase in Child Abuse. The number of fractures due to
nonaccidental causes (child abuse) has risen consistently in the past
decades. In the study of fractures in children ages birth to 3 years
old by Kowal-Vern et al.,⁵² the
number of fractures due to abuse increased almost 150 times from 1984
to 1989. This increase was attributed to a combination of improved
recognition, better social resources, and a true increase in the number
of cases of child abuse.

The anatomic areas most often fractured seem to be the
same in the major series, but these rates change with age. For example,
supracondylar fractures of the humerus are most common in the first
decade, with a peak at age 7. Fractures of the femur are most common in
children ages 0 to 3. Fractures of the physis are more common just
before skeletal maturity. This variation is best illustrated in Cheng
and Shen’s data (Fig. 1-5).²²

Landin’s Age Patterns. Landin found a similar age variability and divided it into six distinct patterns (Fig. 1-6).⁵⁸
When he compared these variability patterns with the common etiologies,
he found some correlation. For example, late-peak fractures (distal
forearm, phalanges, proximal humerus) were closely correlated with
sports and equipment etiologies. Bimodal pattern fractures (clavicle,
femur, radus and ulna, diaphyses) showed an early increase from lower
energy trauma, then a late peak in incidence due to injury from high-
or moderate-energy trauma. Early peak fractures (supracondylar humeral
fractures are a classic example) were due mainly to falls from high
levels.

The overall incidence of fractures occurring because of
play activity in the home environment increases with age. Only 15%
occur in toddlers, but 56% occur during older years.¹³⁴

Early reports of children’s fractures lumped the areas
fractured together, and fractures were reported only as to the long
bone involved (e.g., radius, humerus, femur).¹¹,⁴¹,⁴⁴,⁶²,⁶⁶
More recent reports have split fractures into the more specific areas
of the long bone involved (e.g., the distal radius, the radial neck,
the supracondylar area of the humerus).²²,⁴⁴,⁵⁸,⁹⁷,¹³⁴

In children, fractures in the upper extremity are much more common than those in the lower extremity.⁴¹,⁴⁴
Overall, the radius is the most commonly fractured long bone, followed
by the humerus. In the lower extremity, the tibia is more commonly
fractured than the femur (Table 1-2).

Given the fact that different reports classify fractures
somewhat differently, it is somewhat of a challenge to distill detailed
and accurate prevalence data for specific fractures In trying do so, we
have identified areas common to a number of recent reports,²²,⁴⁴,⁵⁸,⁹⁷,¹³⁴
but have taken some liberties in doing so. For example, distal radial
metaphyseal and physeal fractures were combined as the distal radial
fractures. Likewise, the carpals, metacarpals, and phalanges were
combined to form the region of the hand and wrist. All the fractures
around the elbow, from those of the radial neck to supracondylar
fractures, were grouped as elbow fractures. This grouping allows
comparison of the regional incidence of specific fracture types in
children (Table 1-3).

The individual reports agreed that the most common area
fractured was the distal radius. The next most common area, however,
varied from the hand in Landin’s series⁵⁸ to the elbow (mainly supracondylar fractures) in Cheng and Shen’s series.²¹,²²

The incidence of physeal injuries overall varied from 14.5%²⁴ to a high of 27.6%.⁷² To obtain an overall incidence of physeal fractures, six reports totaling 6479 fractures in children were combined.¹³,²⁴,⁷²,⁷⁹,⁹⁷,¹³⁴ In this group, 1404 involved the physis,

producing an average overall incidence of 21.7% for physeal fractures (Table 1-4).

The overall incidence of open fractures in children is
consistent. The data were combined from the four reports in which the
incidence of open fractures was reported.²²,⁴¹,⁷²,¹³⁴
The incidence in these reports varied from 1.5% to 2.6%. Combined,
these reports represented a total of 8367 fractures with 246 open
fractures, resulting in an average incidence of 2.9% (Table 1-5).

Regional trauma centers often see patients exposed to
more severe trauma, so there may be a higher incidence of open
fractures in these patients. The incidence of open fractures was 9% in
a report of patients admitted to the trauma center of the Children’s
National Medical Center, Washington, DC.¹⁷

Multiple fractures in children are uncommon: the
incidence ranges in the various series from 1.7% to as much as 9.7%. In
four major reports totaling 5262 patients, 192 patients had more than
one fracture (Table 1-6).²²,⁴¹,⁴⁴,¹³⁴ The incidence in these multiple series was 3.6%.

Fractures in Weak Bones. Children with generalized bone
dysplasias and metabolic diseases that produce osteopenia (such as
osteogenesis imperfecta) are expected to have recurrent fractures.

In
these patients, the etiology is understandable and predictable.
However, some children with normal osseous structures are prone to
recurrent fractures for reasons that remain unclear. The incidence of
recurrent fractures in children is about 1%.³²

Landin and Nilsson⁵⁹
found that children who sustained fractures with relatively little
trauma had a lower mineral content in their forearms, but they could
not correlate this finding with subsequent fractures. Thus, in children
who seem to be structurally normal, there does not appear to be a
physical reason for their recurrent fractures.

Failure to find a physical cause for repeat fractures
shifts the focus to a psychological or social cause. The one common
factor in accident repeaters has been a high incidence of dysfunctional
families.⁴⁶ In Sweden, Westfelt
found that children who were accident repeaters came from “socially
handicapped” families (i.e., those on public assistance or those with a
caregiver who was an alcoholic).⁸⁶
Thus, repeat fractures are probably due more to behavioral or social
causes than physical causes. Landin, in his follow-up article,⁵⁷
followed children with repeat fractures (four or more) into adolescence
and adulthood. He found these children had a significantly increased
incidence of convictions for serious criminal offenses when compared
with children with only one lifetime fracture.

Fractures sustained in the home environment are defined
as those that occur in the house and surrounding vicinity. These
generally occur in a fairly supervised environment and are mainly due
to falls from furniture, stairs, fences, and trees.

Falls from Heights. Falls can vary in severity from a
simple fall while running to a fall of great magnitude, such as from a
third story window. In falling from heights, adults often land on their
lower extremities, accounting for the high number of lower extremity
fractures, especially the calcaneus. Children tend to fall head first,
using the upper extremities to break the fall. This accounts for the
larger number of skull and radial fractures in children. Femoral
fractures also are common in children falling from great heights. In
contrast to adults, spinal fractures are rare in children who fall from
great heights.⁸,⁷⁶,¹¹⁴,¹¹⁷
In one study, children falling three stories or less all survived.
Falls from the fifth or sixth floor resulted in a 50% mortality rate.⁸

Social Factors. Interestingly, a Swedish study⁸⁶
showed that an increased incidence of fractures in a home environment
did not necessarily correlate with the physical attributes or poor
safety precautions of the house. Rather, it appears that a disruption
of the family structure and presence of social handicaps (alcoholism,
welfare recipients, etc.) is an important risk factor for pediatric
fracture.

The supervised environments at school are generally
safe, and the overall annual rate of injury (total percentage of
children injured in a single year) in the school environment ranges
from 2.8% to 9.2%.¹⁵,⁶⁰,⁸⁶,¹¹²
True rates may be higher because of inaccurate reporting, especially of
mild injuries. In one series, the official rate was 5.6%, but when the
parents were closely questioned, the incidence of unreported, trivial
injuries was as much as 15%.³⁴ The
annual fracture rate of school injuries is low. Of all injuries
sustained by children at school in a year, only 5% to 10% involved
fractures.³⁴,⁶⁰,¹¹² In Worlock and

Stower’s series of children’s fractures from England,¹³⁴ only 20% occurred at school. Most injuries (53%) occurring in school are related to athletics and sporting events,⁶⁰
and injuries are highest in the middle-school children. The peak time
of day for injuries at school is in the morning, which differs from the
injury patterns of children in general.⁶⁰

Playground Equipment. Play is an essential element of a
child’s life. It enhances physical development and fosters social
interaction. Unfortunately, unsupervised or careless use of some play
equipment can endanger life and limb. When Mott et al.⁸¹
studied the incidence and pattern of injuries to children using public
playgrounds, they found that approximately 1% of children using
playgrounds sustained injuries. Swings, climbers, and slides are the
pieces of playground equipment associated with 88% of the playground
injuries.⁶⁹

In a study of injuries resulting from playground equipment, Waltzman et al.¹²⁸
found that most injuries occurred in boys (56%) with a peak incidence
in the summer months. Fractures accounted for 61% of these injuries,
90% of which involved the upper extremity and were sustained in falls
from playground equipment such as monkey bars and climbing frames.
Younger children (1 to 4 years old) were more likely to sustain
fractures than older children.

Similar observations were made in a study by Lillis and Jaffe⁶³
in which upper extremity injuries, especially fractures, accounted for
most of hospitalizations resulting from injuries on playground
equipment. Older children sustained more injuries on climbing
apparatus, whereas younger children sustained more injuries on slides.

Loder et al.⁶⁴
utlilized the National Electronic Injury Surveillance System (NEISS)
dataset to explore the demographics of playground equipment injuries in
children. Monkey bars were the most common cause of fractures. In
another study looking specifically at injuries from monkey bars, the
peak age group was the 5- to 12-year-old group, with supracondylar
humeral fractures being the most common fracture sustained.⁷⁰

The correlation of the hardness of the playground surface with the risk of injury has been confirmed in numerous studies.⁵⁶,⁶⁴,⁸²,⁸³
Changing playground surfaces from concrete to more impact-absorbing
surfaces such as bark reduced the incidence and severity of head injury
but increased the tendency for long bone fractures (40%), bruises, and
sprains. Chalmers et al.¹⁹ determined that the height of the equipment was just as great a risk factor as the surface composition.

Public playgrounds appear to have a higher risk for
injuries than private playgrounds because they usually have harder
surfaces and higher pieces of equipment,⁹⁰ although playground injury was most likely to occur at school compared to home, public, and other locations.⁹¹

Bicycle Injuries. Bicycle injuries are a significant cause of mortality and morbidity for children.⁹⁶ Bicycle mishaps are the most common causes of serious head injury in children.¹³¹ Boys in the 5- to 14-year age group are at greatest risk for bicycle injury (80%). Puranik et al.⁹⁶
studied the profile of pediatric bicycle injuries in a sample of 211
children who were treated for bicycle-related injury at their trauma
center over a 4-year period. They found that bicycle injuries accounted
for 18% of all pediatric trauma patients. Bicycle/motor vehicle
collisions caused 86% of injuries. Sixty-seven percent had head
injuries and 29% sustained fractures. More than half of the incidents
occurred on the weekend. Sixteen percent were injured by ejection from
a bicycle after losing control, hitting a pothole, or colliding with a
fixed object or another bicycle. Fractures mainly involved the lower
extremity, upper extremity, skull, ribs, and pelvis in decreasing order
of incidence.

Low Helmet Use. More importantly, the study detected
that the use of safety helmets was disturbingly low (<2%). Other
studies confirm the observation that fewer than 13% to 15% of children
wear helmets while riding bicycles.³⁵,¹⁰¹ The Year 2000 Health Objectives called for helmet use by 50% of bicyclists.⁹⁵ Even as recently as 2003, the use of bicycle helmets was still below 20%.⁴²
Research has shown that legislation, combined with education and helmet
subsidies, is the most effective strategy to increase use of safety
helmets in child bicyclists.¹⁸ As
public awareness of both the severity and preventability of
bicycle-related injuries grows, the goal of safer bicycling practices
and lower injury rates can be achieved.⁹⁶

Injuries from Bicycle Parts. Bicycle spokes and handle
bars also are responsible for many fractures and soft tissue injuries
in children. D’Souza et al.²⁹ and Segers et al.¹⁰⁹
found that bicycle spoke injuries are typically sustained when the
child’s foot is caught in the spokes of the rotating wheel. Of 130
children with bicycle spoke injuries, 29 children sustained fractures
of the tibia, fibula, or foot bone. Several had lacerations and soft
tissue defects. D’Souza et al.²⁹
suggested that a mesh cover to prevent the toes from entering between
the spokes and a plastic shield to bridge the gap between the fork and
horizontal upright could substantially decrease the incidence of these
injuries.

Skateboarding. Skateboarding and in-line skating have
experienced a renewed surge in popularity over the past three decades.
With the increasing number of participants, high-tech equipment
development, and vigorous advertising, skateboard and skating injuries
are expected to increase. There was an initial increase in the early
1980s, with a decrease after 1993. Since 1998, there has been an
increase in the number of skateboard injuries.⁷⁵
Because the nature of skateboarding encompasses both high speed and
extreme maneuvers, high-energy fractures and other injuries can occur,
as highlighted by several studies.³⁷,⁸⁸,⁹³
Studies have shown that skateboarding-related injuries are more severe
and have more serious consequences than roller-skating or in-line
skating injuries.⁸⁸ In a study of skateboarding injuries, Fountain et al.³⁷
found that fractures of the upper or lower extremity accounted for 50%
of all skateboarding injuries. Interestingly, more than one third of
those injured sustained injuries within the first week of
skateboarding. Most injuries occurred in preadolescent boys (75%) 10 to
16 years of age; 65% sustained injuries on public roads, footpaths, and
parking lots. Several reports³⁷,¹⁰⁸
have recommended safety guidelines and precautions such as use of
helmets, knee and elbow pads, and wrist guards, but such regulations
seldom are enforced.

Roller Skates and Inline Skates. In a study of in-line skate and roller skate injuries in childhood, Jerosch et al.⁴⁵ found

that in a group of 1036 skaters, 60% had sustained injuries. Eight
percent of these were fractures, mostly involving the elbow, forearm,
wrist, and fingers (78%). Fewer than 20% used protective devices, and
most lacked knowledge of the basic techniques of skating, braking, and
falling. In a larger study of 60,730 skating injuries in children,
Powell and Tanz⁹³
found that 68% of the children were preadolescent boys with a mean age
of 11.8 years. Fractures were the most common injury (65%) and two
thirds of these involved the distal forearm. Two and a half percent
required hospital admissions; 90% of these admissions were for a
fracture. Similarly, Mitts and Hennrikus⁷⁸
found that 75% of in-line skating fractures in children occurred in the
distal forearm as a result of falls on the out-stretched hand. One in
eight children sustained a fracture during the first attempt at the
sport. The orthopaedic community has an obligation to educate the
public on the need for wearing wrist guards when using in-line skates
or roller skates.

Skate Parks Actually Increase the Injury Rate. It was
thought that formal skate parks could decrease the injury rate.
However, a study by Sheehan et al.¹¹¹
demonstrated that dedicated skate parks led to an increase in pediatric
fractures referred to the hospital. The authors suggested that there
should be closer supervision and training of children and more emphasis
on limb protective gear.

Inline Scooters. Since 2000, a substantial increase in
injuries related to nonmotorized scooters (kickboards) has been
observed among children. Most of the scooter-related accidents were
caused by the wheels of the scooter getting caught by uneven ground,
whereas most skateboard accidents occurred during attempted trick
maneuvers. Protective gear was seldom used.²⁰,⁷¹,¹⁰⁵ Scooters seem to have a high incidence of collisions with motor vehicles.⁷¹ The recent motorizing of the scooters will only increase the severity of the injuries sustained.

Trampoline-Related Injuries. Trampolines enjoyed
increasing popularity in the 1990s and are a significant cause of
morbidity in children. Several studies have noted a dramatic increase
in the number of pediatric trampoline injuries during the past 10
years, rightfully deeming it as a “national epidemic.”³⁹,¹¹⁵

Using the NEISS data, Smith et al.¹¹⁵ estimated that there are roughly 40,000 pediatric trampoline injuries per year. Furnival et al.,³⁹
in a retrospective study over a 7-year period, found that the annual
number of pediatric tramopoline injuries tripled between 1990 and 1997.
In contrast to other recreational activities in which boys constitute
the population at risk, patients with pediatric tramopoline injuries
were predominantly girls, with a median age of 7 years. Nearly a third
of the injuries resulted from falling off the trampoline. Fractures of
the upper and lower extremities occurred in 45% and were more
frequently associated with falls off the trampoline. In another
excellent study on pediatric tramopoline injuries, Smith¹¹⁵
found that there was virtually a 100% increase in injuries from 1990 to
1995, with an average of more than 60,000 injuries per year. Younger
children had a higher incidence of upper extremity fractures and other
injuries. In a later study, Smith and Shields¹¹⁶
reported that fractures, especially involving the upper extremity,
accounted for 35% of all injuries. Interestingly, more than 50% of the
injuries occurred under direct adult supervision. More disturbingly,
73% of the parents were aware of the potential dangers of trampolines,
and 96% of the injuries occurred in the home backyard. These
researchers, along with others,³⁹
rightly concluded that use of warning labels, public education, and
even direct adult supervision were inadequate in preventing these
injuries and have called for a total ban on the recreational, school,
and competitive use of trampolines by children.¹¹⁵,¹¹⁶

Skiing Injuries. In a study of major skiing injuries in children and adolescents, Shorter et al.¹¹³
found more than 90% of injured children were boys 5 to 18 years of age.
Sixty percent of the accidents occurred in collisions with stationary
objects such as trees, poles, and stakes. Most injuries occurred in the
afternoon, among beginners, and in the first week of skiing season.
Fractures accounted for one third of the total injuries sustained. The
two main factors implicated in skiing injuries are excessive speed and
loss of control; effective prevention efforts should target both of
these factors.

Snowboarding Injuries. Snowboarding runs a risk similar to skiing. Bladin et al.¹⁴
found that approximately 60% of snowboarding injuries involved the
lower limbs and occurred in novices. The most common injuries were
sprains (53%) and fractures (26%). Compared with skiers, snowboarders
had 2½ times as many fractures, particularly to the upper limb, as well
as more ankle injuries and higher rates of head injury. The absence of
ski poles and the fixed position of the feet on the snowboard mean that
the upper limbs absorb the full impact of any fall. Wrist braces can
decrease the incidence of wrist injuries in snowboarding.¹⁰³
Of some concern, a recent study has shown that rates of snowboard
injuries seem to be rising, while rates of ski injuries have been flat.⁴³

This category includes injuries sustained by occupants of a motor vehicle and victims of vehicle-pedestrian accidents.

The injury patterns of children involved in motor
vehicle accidents differ from those of adults. In all types of motor
vehicle accidents for all ages, children constitute a little over 10%
of the total number of patients injured.⁵⁸,¹⁰⁵
Of all the persons injured as motor vehicle occupants, only about 17%
to 18% are children. Of the victims of vehicle-versus-pedestrian
accidents, about 29% are children. Of the total number of children
involved in motor vehicle accidents, 56.4% were vehicle-pedestrian
accidents, and 19.6% were vehicle-bicycle accidents.³¹

The fracture rate of children in motor vehicle accidents
is less than that of adults. Of the total number of vehicle-pedestrian
accidents, about 22% of the children sustained fractures; 40% of the
adults sustained fractures in the same type of accident. This has been
attributed to the fact that children are more likely to “bounce” when
hit.³¹

Children are twice as likely as adults to sustain a
femoral fracture when struck by an automobile; in adults, tibial and
knee injuries are more common in the same type of accident. This seems
to be related to where the car’s bumper strikes the victim.¹⁷,¹²⁴ Motor vehicle accidents do produce a high proportion of spinal and pelvic injuries.¹⁷

Recreational all-terrain vehicles (ATVs) have emerged as
a new cause of serious pediatric injury. Using the KID dataset,
Killingsworth et al.⁵⁰ showed that 5292 children were admitted to a hospital in 1997 and 2000 (the two years for which KID

data was available) resulting in 74 million dollars in hospital
charges, with rates of hospitalization increasing 80% between these 2
years. In fact, using the Oregon state database, Mullins et al.⁸⁵
showed that the number of patients who sought tertiary care for severe
injuries caused by off-road vehicles doubled over a period of 4 years.
In contrast to other etiologies of injury, children who sustained
ATV-related fractures had more severe injuries and a higher percentage
of significant head trauma, with 1% of these injuries resulting in
in-hospital death. These statisitics point to the failure of voluntary
safety efforts to date and argue for much stronger regulatory control.

According to the 2007 report of the CPSC, serious ATV
injuries in children younger than 16 years requiring emergency room
treatment rose from 146,000 in 2006 to 150,900 in 2007. In their
11-year review of ATV injuries treated at al level 1 pediatric trauma
center, Kute et al.⁵³ determined
that ATV accident-related admissions increased almost five times and
overall fracture number increased four times over the study period; 63%
of the 238 patients sustatined at least one fracture.

In a review of 96 children who sustained injuries in ATV-related accidents during a 30-month period, Kellum et al.⁴⁸
noted age-related patterns of injury. Younger children (≤12 years) were
more likely to sustain an isolated fracture and were more likely to
sustain a lower extremity fracture, specifically a femoral fracture,
than older children. Older children were more likely to sustain a
pelvic fracture. Kirkpatrick et al.⁵¹
expressed concern about the frequency and severity of fractures about
the elbow in their 73 patients injured in ATV accidents between 2001
and 2007: all six open fractures involving the upper extremity involved
the elbow.

The etiologic aspects of children’s fractures are summarized in Figure 1-7 and Table 1-7.

Gunshot or missile wounds arise from objects projected into space by an explosive device. Gunshot wounds have become

increasingly common in children in the United States.¹³⁰
In a sad reflection of the changing times and the newly pervasive gun
culture, firearms are determined to be second only to motor vehicles as
the leading cause of death in youths. In considering the prevalence of
firearms in the United States, it has been estimated that there are
about 200 million privately owned guns in the United States and that
approximately 40% of US households contain firearms of some type.²⁶

Etiology. In two reports from inner-city hospitals in
the United States in the 1990s, most injuries resulted from random
violence to innocent bystanders; the prime example was “drive-by
shootings.”¹²¹,¹³⁰
Few were self-inflicted, either voluntarily or accidentally. In a 1976
report on patients in a relatively rural setting in Canada, almost all
the missile injuries were accidental, having been caused by the patient
or a close friend or relative.⁶¹

In the urban setting, handguns and rifles are the most common weapons.¹²¹,¹²⁵,¹³⁰ In the rural setting, the most common weapon is a shotgun.⁶¹
The firepower of these weapons has changed over the years. In one urban
hospital reporting gunshot wounds from 1973 to 1983, most of the
injuries were from .32- or .38-caliber weapons; only 5% were
high-caliber or high-velocity weapons.⁸⁷
In a later study of gunshot wounds from the same institution from 1991
to 1994, the incidence of injuries from high-caliber and high-velocity
weapons (e.g., .357 magnum, AK-47, and other assault rifles) had
increased to 35%.¹¹³

In the urban setting, the victims’ ages ranged from 1 to 17 years, and most of the injuries were in children aged 12 to 14.⁸⁷,¹²¹,¹²⁵,¹³⁰ In the rural setting, the patients were younger; the average age was 9 years.⁶¹

Of 839 children sustaining gunshot wounds, 274 (32.6%) involved the extremities.⁸⁷,¹²¹,¹²⁵,¹³⁰ Of the gunshot wounds that involved the extremities, 51.3% produced significant fractures.⁶¹,¹²¹,¹³⁰ No single bone seemed to predominate, although most of the fractures were distal to the elbow.⁸⁷,¹²¹,¹²⁵,¹³⁰

Complications of Gunshot Wounds. The two most common
complications were growth arrest and infection. Other complications
included delayed union and malunion. Considering the magnitude of many
of these injuries, the infection rate for extremity wounds was low
(about 7.3%). The type of missile did not seem to have any relation to
the development of an infection.¹³⁰

In Letts and Miller’s 1976 series, one sixth of the patients had some type of growth disturbance.⁶¹
In a third of their patients, the missile was only in close proximity
to the physis, but still appeared to cause a growth disturbance. In a
1995 report by Washington et al.,¹³⁰
the incidence of missiles’ growth arrest was exactly the same; however,
all were a result of a direct injury to the physis by the missile. None
of their patients with growth arrest had proximity missile wounds. The
higher incidence of growth abnormalities in the 1976 series was due to
the larger number of shotgun and hunting rifle injuries, which
dissipate more of their energy peripheral to the missile track.

In two of the studies in which patients were followed closely, all of the fractures ultimately healed.⁶¹,¹³⁰ On the other hand, DiScala and Sege³³
found in their review of children and adolescents who required
hospitalization for gunshot wounds that almost half of them were
discharged with disabilities.

Prevention. In a 1999 report, Freed et al.³⁸
analyzed the magnitude and implications of the increasing incidence of
firearm-related injuries in children. They suggested a product-oriented
approach, focusing on the gun, in an attempt to provide an efficient
strategy of gun control and hence reduce the disturbing trend of
firearm-related injuries and death among youths. Rather than modifying
behavioral or environmental issues, which are more complex, they
suggested focusing primarily on strategies that offset the
accessibility and design of firearms. In brief, these strategies
included reducing the number of guns in the environment through
restrictive legislation, gun buy-back programs, gun taxes, physician
counseling, and modifying the design of guns to make them more
child-proof and prevent unauthorized and unintended use.

Nutrition. In a study in Spain, a significant difference
in fracture rates was found when cities with a high calcium content in
their water were compared with those with a lower calcium content. With
all other factors being equal (e.g., fluoride content, socioeconomic
background), children who lived in the cities with a lower calcium
content had a higher fracture rate.¹²⁶

An increase in the consumption of carbonated beverages
has been shown to produce an increased incidence of fractures in
adolescents.¹³⁵

Bone Density. Bone density may be a factor, but the data are unclear. Landin and Nilsson⁵⁹
found that the mineral content of the forearms was lower in children
who sustained fractures from mild trauma than in children who had never
sustained fractures. It was not significantly different, however, in
those sustaining fractures from severe trauma. This study used
measurements of bone density of the cortical bone in the forearms. Cook
et al.,²⁷ using measurements of bone
density obtained from trabecular bone in the spine and femoral neck,
found no difference between children who had sustained fractures and
those who had not.

Fractures not related to birth trauma reportedly occur
in 1% to 2% of low-birth-weight or premature infants during their stay
in a neonatal intensive care unit.⁶
A combination of clinical history, radiographic appearance, and
laboratory data has shown evidence of bone loss from inadequate calcium
and phosphorus intake in these infants. Correcting the metabolic status
of these low-birth-weight infants, with special emphasis on calcium and
phosphorus intake, appears to decrease the incidence of repeat
fractures and to improve the radiographic appearance of their bony
tissues. Once the metabolic abnormalities are corrected, this temporary
deficiency seems to have no long-term effects. When premature infants
were followed into later years, there was no difference in their
fracture rate compared with that of children of normal birth weight.³⁰