The Epidemiology and Management of 10,486 Children With Fractures in Shenzhen: Experience and Lessons to be Learnt

Purpose: To explore and analyze the causes and related inuencing factors of pediatric fractures, and provide theoretical basis for reducing the incidence and adverse effects of pediatric fractures. Methods: This study retrospectively analyzed the epidemiological characteristics of fractures in children aged ≤ 18 years old who were admitted to the our hospital between July 2015 and February 2020. Results: A total of 10486 pediatric patients were included in the study, of whom 6,961 (66.38%) were boys, and 3,525 (33.62%) were girls. For the fracture incidence, age group of the 3-6 years reached the peak. 5,584 (60.76%) children were operated upon within 12 h after admission. The top three types of fractures were the distal humerus (3,843 sites, 27.49%), distal ulna (1,740 sites, 12.44%), and distal radius (1,587 sites, 11.35%). The top three causes of injury were falls (7,106 cases, 82.10%), car accidents (650 cases, 65.72%), and pinch injuries (465 cases, 5.37%). Fractures predominantly occurred between July and November (4,664 cases, 48.87%) and on Saturdays and Sundays (3,172 cases, 33.24%). The highest number of hospital visits occurred between 20:00 and 00:00 (4,339 cases, 45.46%). Conclusion: We should strengthen preschool children health and safety education and increase protective measures. It is necessary to allocate medical resources in a targeted manner, to establish a closed-loop regional pediatric trauma treatment system centered on trauma centers that links together regional trauma centers, and to establish substantively operating comprehensive pediatric trauma treatment teams that follow a multidisciplinary treatment model.


Introduction
Fractures are very common in pediatric trauma and account for 10-25% of all fractures; moreover, the incidence of fractures has been increasing yearly [1,2,3]. Previous studies have reported the epidemiology of pediatric fractures. For example, from 1993 to 2006 in Malmö, Sweden, the incidence of fractures declined for girls, but not for boys [4]. In addition, data from Switzerland and Norway on long bone pediatric and adolescent fractures showed differences in fracture distribution with respect to sex and age [ 5,6]. A retrospective study conducted by Rennie et al. in Edinburgh in 2000 showed a yearly incidence of 20.2 pediatric fractures per 1000 children [7]. The pediatric fracture sites also varied by country and region [8,9]. A recent study conducted in India showed that elbow fractures are the most common pediatric fracture types [10], which was similar to the results of a study conducted in Hong Kong, China [11]. According to the 2015 Chinese National 1% Population Sample Survey, the population of children aged 0-17 years in China was 271 million, which accounted for 12.9% of the global pediatric population. With the development of urbanization, motorization, and the construction industry in China, the pattern of post-traumatic fractures in children and adolescents is likely to change. However, in China, no systematic studies have been conducted on the overall pattern and epidemiological trends of pediatric and adolescent fractures.
Sports injuries were divided into slide falls, trampoline falls, single/parallel bar falls, skateboard falls, tness equipment falls, basketball falls, rocking horse falls, falls while running, balance bike falls, ice skating falls, swing falls, ski falls, taekwondo falls, kick injuries, dance falls, playground falls, soccer falls, falls during physical education activities, jump rope falls, swimming-related injuries, martial arts falls, and others.
Based on the geographic area where the fracture occurred, we divided the administrative areas of Shenzhen into four regions: eastern Shenzhen (Longgang, Yantian, Pingshan, and Dapeng New Districts), western Shenzhen (Baoan, Guangming, and Nanshan Districts), northern Shenzhen (Longhua District); and southern Shenzhen (Futian and Luohu Districts). Other regions involved were Huizhou City, Guangdong Province; Dongguan City, Guangdong Province; and Hong Kong Special Administrative Region, China. This study was approved by the Ethics Committee of our hospital.

Results
Age and sex A total of 10486 pediatric patients were included in the study, with 6,961 (66.38%) boys and 3525 (33.62%) girls. Boys had more fractures than girls in all age groups. The age groups with the highest and lowest number of fractures in boys were the preschool children and infant groups, respectively; whereas those in girls were the preschool children and adolescents groups, respectively.
Among the four patient age groups, the infants group had the lowest number of fracture cases (1,227 cases, 11.70%), including 694 (56.56%) boys and 533 (43.44%) girls. Due to the rapid growth and development of children, the number of fracture cases gradually increased and peaked in the preschool children group (4,380 cases, 41.77%), including 2,719 (62.08%) boys and 1,661 (37.92%) girls. Thereafter, the number of fracture cases showed a decreasing trend, with 3,622 cases (34.54%) in the school children group, including 2,478 (68.42%) boys and 1,144 (31.58%) girls. The number of fracture cases in the adolescent group was 1,257 (11.99%), including 1,070 (85.12%) boys and 187 (14.88%) girls. Moreover, the sex ratio (males to females) for fractures increased with age, reaching a peak of 5.7:1 in the adolescents group. (Table 1) Tables 2 and 3) Days and months of fracture occurrence By retrospectively analyzing the 9,544 pediatric patients admitted to our hospital over 4 calendar years from January 2016 to December 2019, we found that the peak time for fractures was on Saturdays and

Discussion
The present study revealed many salient ndings. A) First, more boys than girls were hospitalized for pediatric fractures, and the highest number of fractures occurred in children aged 3-6 years. B) Second, the most common fractures involved the distal humerus and resulted from falls; moreover, the most common epiphyseal injury involved the distal radius. C) Third, the most common concomitant nerve injury was radial nerve injury; furthermore, the most common concomitant multisystem injury was nervous system injury. D) Fourth, 11.40% of pediatric patients had two or more concomitant fracture sites, and 2.46% had multisystem injuries, including 23 patients with concomitant shock. E) Fifth, the treatment of fractures mainly involved surgical treatment by closed reduction. F) Sixth, of the 9,191 patients in the surgical treatment group, 5,584 received surgical treatment within 12 h of admission. G) Lastly, fractures occurred more frequently between July and November, and on Saturdays and Sundays. The peak hours of admission were from 20:00 to 00:00, and 7.30% of patients visited our hospital visit > 72 h post-injury. Between 20:00 and 00:00, 86.06% of patients were referred to our hospital after their initial visit at a local hospital.
More educational programs on safety measures should be organized A study by Rennie et al. [7] found that fractures in children aged 0-16 years occurred most frequently between the ages of 5-11 years, which accounted for 51.3% of fractures in all age groups. However, the Chinese scholars Chen et al. [12] found that the highest incidence of fractures occurred at approximately 3 years of age. This is consistent with our ndings on the age distribution of pediatric fractures wherein fractures occurred most frequently between the ages of 3-6 years (41.77%), followed by 7-11 years (34.54%). The discrepancy in the peak age group for fractures between China and other countries may be related to differences in the use of sports protective equipment and health education. Our ndings suggest the need for improvements in the abovementioned areas for children in fracture-prone age groups.
Landin [13] showed that during the period from birth to 16 years, the incidence of fractures was 42% for boys and 27% for girls. Our study also found a signi cantly higher proportion of boys than girls. Hedström et al. [3] and Cooper et al. [14] found that the peak age for fractures in girls was 11-12 years, whereas that for boys was 13-14 years; however, the peak for fractures in our study, for both boys and girls, was 3-6 years. Rennie et al. [7] showed that the incidence of fractures increased with age in both boys and girls before the age of 11 years; nonetheless, the incidence in girls began to decline after the age of 5 years, whereas that in boys continued to increase and was about twice that of girls by the age of 13 years. Our ndings corroborated with those of Rennie et al., as we found that the number of fractures increased in boys but decreased in girls after 11 years, with a male to female ratio of 5.7: 1 at 11 years. Thus, despite their larger and stronger bones, boys have a higher incidence of fractures than girls, which may be attributed to their preference for more intense, competitive, and confrontational activities.
Therefore, health education should place more emphasis on safety education for boys, especially those who are active and like to participate in high-risk sports that can lead to serious injuries; in addition, more attention should be placed on children aged 3-6 years.
Some studies have suggested that major pediatric fracture sites vary with age. More speci cally, clavicular, distal humeral, distal radial, and metacarpal fractures are the most common in children aged < 1 year, 1-3 years, 4-14 years, and 15-16 years, respectively [9]. However, in the present study, distal humeral and distal radial fractures were most common in children aged < 11 years and 12-18 years, respectively.
Brudvik et al. [15] showed that pediatric fractures occur primarily in the upper limbs with the distal forearm as the most common fracture site. Children have a habit of extending their arms to protect themselves when they fall, which is the main reason why they are more prone to having upper limb fractures [11]. Fractures of the phalanges of ngers are a common type of pediatric fractures resulting from trauma in emergency departments and outpatient clinics, with a high annual incidence [18][19][20][21]. A survey found a very low annual incidence of hand fractures among toddlers in the United Kingdom (34/100,000 children); nevertheless, this gure increases by approximately 20-fold after the age of 10 years, reaching 663/100,000 in children aged 11-18 years [21]. Many previous studies revealed that hand fractures were more common in boys than in girls, with 65-75% of fractures occurring in boys and a peak occurring around 9-14 years [22][23][24]. However, among the 1,112 cases of fractured phalanges of ngers in this study, which included 696 boys and 416 girls, the peak and trough ages of fracture occurrence were 3-6 years (469 cases) and 12-19 years (120 cases), respectively. This nding differed from that previously reported in the literature in other countries.
Epiphyseal fractures are a unique type of pediatric fractures, accounting for 15-30% of all pediatric fractures. The incidence of epiphyseal fractures is higher in boys than in girls, and is more common in the upper limbs [25,26]. Komura et al. [27] found that, of all epiphyseal fractures, those of the distal radius are relatively common. Steinberg et al. [28] showed that about 50% of epiphyseal fractures occurred in the proximal radius. Peterson et al. [29] found that the most common site of epiphyseal fracture was the phalanges of the ngers. In the present study, epiphyseal fractures occurred in 1,209 cases, accounting Although there are multiple causes of pediatric fractures, falls during normal playing and sports represent a major cause. Both Gouiding et al. [30] and Mansoor et al. [31] identi ed play-related falls as a primary cause of injury. Similarly, our study identi ed falls as the most common cause of injury, accounting for 67.77% of all causes. It is also the most common cause of injury in all age groups, with the highest number of reported cases of injury in children aged 3-6 years. Children in this age group are able to move independently, and often trip and fall while playing.
A study conducted by Osmond et al. [32] concluded that road tra c injuries represented a common cause of severe trauma in children, and this also varied across age groups. In this study, road tra c injuries included 650 and 211 cases involving car accidents and bicycle falls, respectively. Car accidents peaked at 3-6 years (328 cases), whereas bicycle falls peaked at 7-11 years (85 cases). Such high-energy traumas often lead to open fractures, polytrauma, and shock, which pose a serious threat to children's health. Therefore, parents, the community, and schools, should be educated on tra c safety, including the use of car safety seats, wearing protective gear while riding bicycles, and increased supervision [33,34].
With respect to bicycle-related fractures, attention should also be given to bicycle-spoke injuries, which occur when the limb comes into contact with the spokes of the bicycle wheel, thereby leading to entanglement of the limb in the spokes and crushing of the limb against the bicycle frame. Injuries are usually sustained by children who are bicycle passengers, and involve mainly lower limb soft tissue damage, followed by lower limb fractures [35,36]. In this study, 103 cases of bicycle-spoke injury were observed, predominantly in children aged 3-6 years, and the most common fracture site was the distal tibia. Although bicycle-spoke injuries are usually not life-threatening, the resulting socioeconomic damage is substantial. We believe that, in order to prevent bicycle-spoke injuries, safety education for guardians should be vigorously promoted, as this enhances supervision and prevention. With the growing popularity of bicycle sharing, bicycle manufacturers should improve the design and structural defects of the bicycle itself. This can be done by installing protective equipment (for instance, protective nets) to preventing toes from getting caught between spokes and designing special safety child backseats.
In this study, falls from height affected 114 patients, and peaked in children aged 3-6 years (50 cases).
We believe that emphasis should be placed on the protection of high-rise buildings, and promoting safety education for guardians of children in the peak age group of 3-6 years. For example, in New York, a program themed "Children Can't Fly" was organized to prevent fall injuries. Through extensive public safety health education, the installation of guardrails on high-rise windows, and the installation of protective nets, remarkable results were achieved in the number of fall injury victims, which dropped by 50% after 3 years, and by 96% after 7 years [37].
The third leading cause of injury in this study was pinch injuries (465 cases). Door-pinch injuries were observed in 403 patients, which primarily resulted in fractures of the phalanges of ngers. Al-Anazi [38] and Doraiswamy [39] concluded that door-related nger crush injury is the leading cause of nger injury in children. These injuries frequently occur at home, where the affected ngers are often crushed against the hinged side of the door, and are more likely to occur in younger children. The majority of door-pinch injuries occur in the presence of an adult, thus highlighting the need for preventive measures. We should increase guardians' awareness of these injuries and provide more educational programs on safety measures. Moreover, since ngers are most likely crushed in the hinged side of the door, nger guards can be installed to prevent nger entrapment. Furthermore, triangular rubber and plastic or wooden door wedges can be inserted into the bottom of the door to prevent its automatic closure.
There is a distinct seasonal pattern in the occurrence of pediatric fractures. Sinikumpu et al. [40] showed, in a study on fractures and weather, that most fractures occurred on dry days (79.7%) as opposed to rainy days, with a 3.5-fold higher risk of fracture on dry days. In a study conducted in Ireland, Masterson et al. [41] also showed a signi cant positive correlation between the number of sunshine hours per month and the corresponding number of hospital admissions for fractures, as well as a weak negative correlation between the number of fractures and the amount of rainfall per month; similar studies have been conducted in countries with summer holiday customs [42]. Fractures predominantly occurred in the afternoon, possibly due to the after-class or after-school hours during which schoolchildren are prone to fractures [43]. The high incidence of fractures in summer and autumn, and the lower incidence in winter, were also con rmed in the present study. This may be related to the favorable climate and temperature in summer and autumn, the long sunshine hours, the increased outdoor activity of children, and the use of thinner and fewer clothing that neither absorbs collision energy nor provides cushioning in the event of an accident. In addition, our study showed that July to November were the peak months for fractures. Apart from the aforementioned reasons, this may also be related to the National Day holiday and summer vacation in China, during which children spend more time outdoors and are exposed to accidental injuries and fractures. Therefore, on the one hand, medical resources need to be increased to ensure the priority and quality of care for children with fractures during this peak period. On the other hand, health education and promotion should be carried out in conjunction with schools and communities prior to this peak period.
A closed-loop regional pediatric trauma treatment system should be established This study documented that in 1,373 (13.09%) patients, the interval between injury and visit to our hospital was greater than 24 h, of which 765 cases (7.30%) had an interval > 72 h. According to the detailed medical records, 606 (79.22%) patients were transferred to our hospital due to unsatisfactory results of inpatient treatment at their local hospitals, none of which were children's specialized hospitals. Disparities exist in the treatment plans for fractures in children and adults, and our ndings suggested that more children's specialized hospitals or PTCs in general hospitals should be established in the region, in order to provide more timely and effective treatment plans for children with fractures.
In addition, this study found that hospital visits for pediatric fractures were mainly concentrated between 20:00 and 00:00 (4,339 cases, 45.46%), and included 3,734 cases of referrals to our hospital after their initial visit to other hospitals. These were all patients who were referred to our hospital for fracture treatment after their initial visit to their local non-specialty hospitals or general hospitals without PTCs. Moreover, the geographical areas wherein the injuries occurred were mainly concentrated in the western and eastern parts of Shenzhen. In response to this peak period, our hospital routinely deploys more doctors and nurses on duty during this time, allocates medical resources such as consumables and beds in advance, and actively coordinates the work of the emergency surgery and surgical anesthesiology departments.
This phenomenon further illustrates the necessity to establish a pediatric trauma treatment network or trauma medical alliance in Shenzhen in particular, and in China as a whole. A closed-loop regional pediatric trauma treatment system centered on trauma centers should be established that can link together the regional trauma centers. In order to promote the normalized and standardized development  [44][45][46][47][48][49]. A PTC classi cation system should be constructed based on the existing hospital grading system combined with regional characteristics. This will help to unify the standard system for trauma center construction, enable the stipulation of systematic provisions for the responsibilities of trauma centers at all levels, and foster the relationship among them. On the one hand, this will equalize the medical resources of each region, whereas on the other hand, it will be possible for children to receive timely treatment in their vicinity. This will result in a joint construction of a pediatric trauma treatment system characterized by "vertical links, initial visits at the primary care level, two-way referrals, and separate treatments for acute and chronic diseases".
A comprehensive pediatric trauma treatment teams that follow a multidisciplinary treatment model should be substantively operated In addition to injuries to the musculoskeletal system, there were 258 patients with concomitant injuries to other systems, most commonly involving neurological, respiratory, and digestive injuries. More so, it is challenging to treat high-energy injuries such as road tra c injuries and fall injuries. Twenty-three patients in our study had concomitant shock. An effective trauma care system ameliorate the prognosis of patients with trauma more than the individual clinical experiences of physicians. The ndings of a meta-analysis showed that a well-developed emergency trauma system reduces overall trauma mortality by 15% [50]. An integrated model of care is best suited to children with multiple injuries, as it provides a stable and specialized trauma team responsible for every aspect of care [51]. The establishment and volition of the American trauma system over the past 40 years has achieved the standardization of trauma care through trauma centers, which has greatly helped reduce the impact of trauma on people of all ages [52]. We believe that we should learn from the successes of European and American countries in order to strengthen the construction of the pediatric trauma treatment system, the core of which should be the hierarchical construction of trauma centers, with an emphasis on the multidisciplinary integrated treatment of polytrauma and critical trauma patients. We should establish a substantively operating comprehensive pediatric trauma treatment team, which consists of xed personnel from core and supporting specialties. Core specialties should include orthopedics, general surgery, neurosurgery, urology, thoracic surgery, emergency medicine, anesthesiology, intensive care medicine, medical laboratory, and medical imaging; whereas supporting specialties should include pediatrics, burns department, cardiovascular surgery, plastic surgery, ophthalmology, otorhinolaryngology, stomatology, psychiatry, rehabilitation medicine, medical imaging, blood transfusion, etc. The biggest advantage of this approach is that the information platform, treatment process, sta ng, and equipment supply are based on the principles of fastest e ciency and optimal process. This eliminates the previous practice of temporarily inviting relevant specialties for consultation in response to the patient's condition and evolution, thus greatly reducing the rescue time and improving the treatment outcome. Furthermore, among the 9,191 patients who underwent surgical treatment, 5,584 (60.76%) were operated upon within 12 h of admission, which shows that the substantively operating comprehensive pediatric treatment team implemented by the our hospital has begun to bear fruit. This process speci cally involves dispatching personnel from the orthopedic department to set up an emergency pediatric fracture specialty in the emergency department for the initial diagnosis and treatment of pediatric patients, setting up an inpatient department with beds, working with other departments of pediatric medicine and pediatric surgery to improve the relevant treatment plan, and coordinating with the surgical anesthesiology department for pediatric patients with indications for surgery in order to complete the relevant pre-operative preparations within 12 h.

Conclusions
With regard to the children population, we should strengthen health and safety education for children, parents, communities, and schools in accordance with their age, sex, cause of injury, fracture site, and other characteristics; increase protective measures for children's activities; and prevent pediatric fractures through government legislation and other measures. Moreover, a multidisciplinary treatment model based on PTCs should be established. On the one hand, the government, industry associations, and hospitals at all levels should promote the construction of trauma centers, establish a pediatric trauma treatment network or trauma medical alliance, and establish a closed-loop regional pediatric trauma treatment system centered on trauma centers that can link together the regional trauma centers. On the other hand, a substantively operating comprehensive pediatric trauma treatment team should be established, which follows the treatment process of multidisciplinary cooperation, in order to reduce the mortality and disability rates of children with trauma. This will positively in uence the protection of lives and ameliorate the health of children. was waived because of the register design of this study, which did not involve any additional risk for patients. The need for written informed consent was waived by the Shenzhen Children's Hospital ethics committee due to retrospective nature of the study.

Consent for publication
Not applicable.

Availability of data and materials
The data sets used and analysed during the current study are available from these corresponding authors on reasonable request.

Competing interests
For all authors, none was declared.
Funding Figure 2 The epidemiology of traumatic fractures in all age range groups This picture shows the characteristics of common fracture sites in the infants group, preschool children group, school children group and adolescent group.