| | (iv) The response of children to traumaAbstract Children differ markedly from adults in their response to trauma. They sustain predictable injuries from typical mechanisms based upon their age and activity profile. They have been shown to have different patterns of injury when sustaining single bone trauma, pelvic fractures and spinal injuries. The microscopic structure of a child's bone is different creating an environment for unique fracture types including those of the physis. The biological healing response differs and children often have the capacity for significant remodelling. Due to the unique anatomy and physiology of a child, the management of severe trauma and multiple injuries is different to that of an adult. Continuing growth and development always needs to be taken into account and will have a bearing on the management plan. The long-term outcome, from both a functional and psychological standpoint, is of great concern as these patients have potentially so much time ahead of them. It is important to be aware of all these differences to ensure that assessment, diagnosis, initial and definitive management, rehabilitation and follow up are optimal for each injured child. Epidemiology  Preventable injury is the leading cause of death in children >1 year of age, most commonly due to pedestrian road traffic injuries. This is followed by non-pedestrian road traffic injuries, suffocation and assault/homicide, fire and falls.1 The proportion of all deaths caused by injury increases with age so that the group at greatest risk are adolescents, although the adolescent death rate due to injury is still only one third of that of adults. Over the last 30 years the death rate due to injury has steadily declined, most likely due to the successful implementation of wide ranging preventative programs. The childhood mortality rate attributable to injury in Scotland has dropped below 10% for the first time.1 Mortality, however, is a poor indicator of the problem because it does not reflect the high ratio of injuries to deaths. It is estimated that for every child killed in a motor vehicle accident, 48 were injured.2 In the first 10 years of life, 85% of children will sustain at least one injury that requires medical treatment. Repeat injury is common with an incidence of 73%.3 20–25% of children are injured each year in the United States of America and Canada. The estimated cost to these societies is $347 billion per year.4 Boys sustain significantly more injuries than girls. Furthermore, they tend to suffer more serious injuries and have a greater subsequent rate of hospitalization. The ratio of injuries in boys to girls also increases with age. The highest rate of injury and mortality is therefore in the adolescent male group. Multiple fractures in children are not common with an incidence of 3.6%.5 Mechanism of injury It is estimated that 10–25% of childhood injuries are fractures. More common non-fatal childhood injuries include dislocations and sprains, open wounds, burns, poisoning, head injury and foreign bodies.3 The most common cause for paediatric fractures is a fall from below bed height and in decreasing incidence other causes are blunt trauma, falls from above bed height, sports, motor vehicle accidents and falls down stairs or slopes. The majority of children's fractures occur in the afternoon or early evening and of these, one third occur at home. Mechanism of injury and age are commonly linked. Injury by falling from monkey bars is most common in the 5–10 year age group. Skateboards play their part in the injury profile of children between 10 and 15 years of age. Sports injuries generally increase in incidence with increasing age to a peak at secondary school age. It tends to be informal games that result in the highest rate of injuries. Supervised sports provide half the injury rate than that of unsupervised sports.6 While team sports such as soccer, rugby and ice hockey are known to provide a steady stream of fractures to the emergency department, these are generally at the less severe end of the spectrum. Serious injuries requiring hospitalization are much more likely to come from skiing, snowboarding, trampolining and cycling. The incidence of trampolining injuries has increased dramatically in the last 15–20 years with the popularity of recreational trampolines. More than 80% of injuries occur at home or a neighbour's house with more than one child on the trampoline at the same time, and by a fall onto the mat. The Canadian Academy of Sports Medicine and Canadian Paediatric Society have reviewed the published literature on the topic and have issued a joint statement recommending that trampolines not be used for recreational purposes nor be regarded as play equipment.7 Children are nearly twice as likely to be involved in motor vehicle accidents as pedestrians than as vehicle occupants. Less than 20% of those injured as vehicle occupants are children.5 Rennie observed that in her series of over 2000 paediatric fractures seen in a single hospital in Edinburgh, Scotland in one year, only four fractures were caused from being the occupant in a motor vehicle accident concluding that car safety and speed control measures may have had an influence in keeping down this particular statistic. This and other similar studies may help to direct the focus of injury prevention programs towards the protection of children as pedestrians where significant improvements can still be made. As a pedestrian, children are less likely to sustain a fracture than adults, perhaps due to the increased musculoskeletal elasticity which serves to dissipate the kinetic energy transferred to the child's body when hit. Due to the point of impact of the bumper on the patient, a fractured femur is more common in children as opposed to adults where a fractured tibia and knee injuries are more common.5 Although spinal and pelvic fractures are generally uncommon in children, a pedestrian road traffic accident is the most common mechanism of injury implicated. Non-accidental injury must always be suspected in cases where multiple injuries are identified in young children if there is no obvious, witnessed explanation for the trauma. Although children of any age can be abused, younger children are more frequently victims. Child abuse has been found in up to half of all children with fractures in the first year of life and in one third of children younger than 3 years of age with a fracture. Rarely are there the so-called pathognomonic signs of corner fractures and multiple fractures at different stages of healing present. The most common fracture seen in child abuse is a single transverse fracture of the femur or humerus. Also of note is that soft tissue injuries are more common than bony injuries. A detailed description of this important topic is beyond the scope of this article, but on the basis that 1–1.5% of all children are abused each year,8 it always warrants consideration. Patterns of injury  Children are much more likely to sustain isolated injuries from relatively low energy mechanisms, but on the occasion when one is presented with a child having been involved in severe trauma, a knowledge of their specific patterns of injury is essential. Limb fractures Rennie described the epidemiology of all paediatric fractures presenting to a single hospital in Edinburgh, Scotland in 2000. 2198 fractures were identified. She found that the overall incidence of fractures was 20.2 fractures/1000/year. 61% were male. 82% of fractures were in the upper limb, 17% were in the lower limb and 0.5% were in the pelvis or spine. Less than 1% of fractures were open and 15% were physeal.9 These results are in line with those previously reported from groups in England, Greece, Scandinavia and the USA, although most report a slightly higher incidence of open injuries (2.9%) and physeal injuries (21.7%).5 The anatomic site of fracture varies with age. In children overall, fractures of the distal radius and ulna are most common, followed by fractures of the hand, elbow, clavicle, radial shaft, tibial shaft, foot, ankle, femur and humerus (excluding distal) respectively. In the 1–3 year age group, however, fractures of the distal humerus predominate, and in 15–16 year old age group, fractures of the metacarpals and phalanges are most common.9, 5 Pelvic fractures Paediatric pelvic fractures are rare and as such the literature is limited. In contrast to adult pelvic fractures, the paediatric pelvic fracture tends to be more stable, has less associated severe haemorrhage, a lower mortality, a lower incidence of associated genitourinary tract injury and is more amenable to non-operative management due to the often stable nature of the injury.10 The child with a pelvic fracture is usually multiply injured. The explanation being that the child's pelvis is inherently more elastic than that of an adult and thus more energy is required to cause bony injury, and should such a force be applied to cause a pelvic fracture, then it is likely that other organ systems will also be injured. The mortality of children with a pelvic fracture is less than 10%.11 Pelvic fractures themselves do not contribute significantly to the overall mortality in a child. Mortality is governed by the associated concomitant injuries, in particular head injury. One reason for this difference in mortality is that the mechanism of injury in the child differs from that of an adult. Life threatening haemorrhage after a pelvic fracture in an adult is commonly due to vascular injury associated with the AP compression injury. This injury configuration also leads to an increased intrapelvic volume in which to exsanguinate. This is a rare mechanism of injury in the child with the vast majority of injuries conforming to a lateral compression type, which is not associated with such severe haemorrhage nor an increased intrapelvic volume. The cause is most commonly a pedestrian road traffic accident, accounting for the lateral compression injuries, followed by a fall from a height. When considering paediatric pelvic fractures, one needs to be aware that the fracture pattern changes with skeletal maturity. The status of the triradiate cartilage is the most useful radiographic feature in this regard. The triradiate cartilage tends to close at age 12 in girls at age 14 in boys. With an open triradiate cartilage the injury patterns consist mainly of isolated pubic symphysis fractures and iliac wing fractures. Acetabular fractures and diastasis of the symphysis pubis and sacroiliac joints are rare in the skeletally immature group. This is accounted for by the fact that joints of the pelvis are more flexible and elastic at this age allowing for greater deformation before permanent diastasis. This also allows a single fracture of the ring as opposed to the double break concept in the adult pelvis. In the child with a closed triradiate cartilage, acetabular fractures and diastasis of the symphysis pubis and sacroiliac joints are much more common.11 The management plan is directly guided by this knowledge, as the group with an open triradiate cartilage typically have stable injuries that respond well to non-operative management whereas the group with a closed triradiate cartilage have unstable patterns of injury that will usually require operative management following the principles set out for adults. Spinal fractures Paediatric spine fractures are also rare comprising a smaller proportion of all fractures when compared to adults, but with head injury aside are associated with the highest mortality of any orthopaedic injury in the child. They should be considered life threatening as they often have severe associated injuries to the head, thorax, abdomen or pelvis.12 When children sustain spinal fractures they tend to be associated with an overall higher injury severity than in the adult due to the more violent mechanism required to produce injury to the childhood axial structures. Spinal fractures are present in approximately 3% of children admitted to hospital with major trauma, and in 10% of cases fractures are present at multiple levels.13 Boys sustain more spinal injuries than girls at all ages, with the biggest difference between the genders occurring in adolescence. This is attributed to a higher participation of boys in contact sports and other risk-taking behaviour. Most spinal fractures occur after road traffic accidents, followed by falls, sporting injuries and assaults. Road traffic accidents account for the majority of severe spinal injuries. As the child increases in age and reaches adolescence, sports injuries play a larger part in the cause of spinal injuries, however, they still tend to account for the less severe injuries in the spectrum seen. Spinal injuries in the child differ markedly from those in the adult population. Children's spinal injuries, in common with pelvic fractures, can be divided into two age groups based on musculoskeletal maturity. The time at which the spine takes on the anatomical and biomechanical characteristics of the adult spine is at 8 years of age. Below the age of 8 years classic paediatric type injuries are much more common with the risk of spinal cord injury being twice as likely.14 Unique to the paediatric population is the phenomenon of SCIWORA (spinal cord injury without radiological abnormality). The term SCIWORA was first used in 1982 in an era prior to the widespread use of Magnetic Resonance Imaging (MRI) and so really referred to a spinal cord injury without associated fractures or dislocations visible on radiographs or computerized tomography (CT) scans. The routine use of MRI will now show abnormality in nearly all cases, perhaps making the use of the term SCIWORA outdated, however it remains a concept and condition of great importance. SCIWORA usually follows high-energy trauma causing a stretch or distraction injury to the relatively flexible spine and accounts for up to 20% of childhood spine injuries. The neurological injury may be complete or incomplete and there are often associated injuries. In terms of vertebral levels injured, cervical spine injuries markedly predominate in all age groups although the dominance of cervical spine injuries gradually decreases with increasing age. In the child younger than 8 years, 75% of cervical spine injuries occur in the upper region, due to their unique anatomic features including a proportionally large and heavy head, whereas 25% of injuries occur in the lower cervical spine. In this age group, thoracic spine injuries are less common than in older children. As the child ages the injuries move caudally so that in adolescents, whilst cervical injuries still predominate overall, lower cervical spine injuries are actually more common (60%) than upper cervical spine injuries (40%) and thoracic spine injuries are more common than in children under the age of 8 years.14 Structural characteristics and related injuries  To understand the nature of children's trauma, one must have a detailed appreciation of the unique biological and structural differences between the skeleton of a child and that of an adult. In this way injury patterns can be predicted and more easily identified, appropriate management plans be devised and a sensible follow up and rehabilitation instigated. The immature skeleton offers some real advantages in the management of trauma, in particular its ability to remodel many fractures to avoid long-term deformity that would be inevitable in the equivalent treatment in the adult. There are also disadvantages and one also has to be mindful that the child's musculoskeletal system is a work in progress, not all bones and physes are the same, and consideration must be given to the potential impact of an injury on the future growth and development of the individual. The fundamental difference between the bone of an adult and a child is that the adult bone has ceased longitudinal growth and is slowly remodelling in response to the extrinsic and intrinsic forces applied to it. The child's bone, on the other hand, is also dealing with the extrinsic and intrinsic forces applied to it, but in a far more dynamic, biologically active fashion whilst also being engaged in longitudinal and appositional growth. The fracture characteristics of a particular region of bone have much to do with the macroscopic and microscopic make up of the tissues from which it is constructed; the composition of fibrous tissue, cartilage, trabecular bone, cortical bone and its haversian systems and pores and the size of the medullary cavity. As the bone grows the density of the cortex increases, particularly in the diaphysis. The cross-sectional porosity of a child's bone is much greater; this may affect the fracture pattern by preventing comminution, which is rarely seen in the child. Adult bone usually fails in tension, whereas a child's bone may fail macroscopically in either tension or compression.15 The diaphysis and the greenstick fracture The child's diaphysis gradually changes with age. The progressive development of Haversian systems and varying concentrations of lamellar and laminar bone make the bone more elastic in nature and combined with a thick periosteum create the environment for a greenstick fracture to occur. A greenstick fracture is characterized by progressive plastic deformation of the bone followed by complete failure of the cortex and periosteum under tension, whereas the cortex and periosteum under compression remain macroscopically intact (Figure 1). Due to the residual plastic deformation, it is common to require “completion” of the fracture by reversing the angulation to allow satisfactory reduction. These fractures are generally stable after reduction due to the periosteum, which can be used to significant advantage by acting as a tension band under a well-moulded cast. With growth the diaphysis of the bone matures before the metaphysis and subsequently becomes more stiff leaving the metaphyseal region vulnerable to greenstick fracture for longer than the diaphysis. The metaphysis and the torus fracture The metaphysis is the most biologically active region of the bone subsequently having a different structure than diaphyseal bone. The metaphysis is composed of thin and fenestrated cortical laminar bone and considerable endosteal trabecular bone. The important biological activity of rapid remodelling leaves the metaphysis vulnerable to injury, as the bone has no time to structurally respond to the forces placed upon it, a luxury afforded to the diaphysis. The torus (or compression) fracture is common at the junction of the woven bone of the metaphysis and lamellar bone of the diaphysis, where the stiffer diaphysis acts as a stress riser. This type of compression fracture is unique to children, and occurs because of greenstick bending and microfracture. Torus fractures actually represent a spectrum of injury from one barely perceptible on a radiograph through to a full greenstick fracture with tension failure of one cortex (Figure 2). Clinically it is important to distinguish true compression fractures as these rarely need reduction, do not displace and heal with rapidity under minimal splintage. The greenstick fracture is not quite so benign and can require reduction, moulded cast, may redisplace and often requires longer periods of immobilization. The physis and the physeal injury Approximately 20% of children's fractures are physeal in nature, however growth disturbance is relatively rare occurring in less than 10% of all physeal injuries. A knowledge of physeal anatomy and the characteristics of physeal injuries will help the clinician predict which injuries are likely to require close observation and subsequent treatment, and which can be safely discharged. The function of the long bone physis (growth plate) is to coordinate rapid, structured, longitudinal growth of the bone, and to a lesser extent, growth circumferentially at the level of the physis (Figure 3). As with most areas of specialized tissue the focus on biological activity is at the expense of some structural integrity, hence leaving the growth plate vulnerable to injury. The physis of the long bone can be considered a three-dimensional disc of tissue between the epiphysis and metaphysis arranged in columnar zones of cells. These zones are traditionally characterized by function, although the reality is that there is a gradual transition from one zone to the next. The zone abutting the epiphysis is the germinal zone followed, in the direction of epiphysis to metaphysis, by the zones of differentiation, proliferation, hypertrophy, provisional calcification, which then transitions into the primary spongiosa of the metaphysis with its looping cascades of capillaries. Each zone has a different mechanical strength and biological activity. In terms of structure, at the level of the germinal zone there is abundant extracellular matrix acting as a scaffold protecting this important nest of cells, however, as one passes toward the metaphysis the proportion of matrix to active cells of the physis decreases. The hypertrophic zone, as one might expect, has the largest cells and therefore the least matrix, leaving it vulnerable to injury. It is particularly susceptible to shear, torsional and tensile forces, but is relatively resistant to compression. Immediately on moving to the zone of provisional calcification, strength increases again due to the appearance of calcium, affording some protection from shear. This further concentrates forces at the zone of hypertrophy abutting it on its epiphyseal side. This is the reason that most physeal injuries occur through the zone of hypertrophy, and why growth disturbance is relatively uncommon. Injury to the germinal zone is much more likely to result in a growth disturbance due to it being the reservoir of cells programmed to enter the process of growth and development. The job of circumferential physeal growth and further structural support to the physis is undertaken by specialized areas at the periphery of the physeal disc. The zone of Ranvier has a triangular cross-sectional shape and runs around the periphery of the physis at its junction with the metaphysis. It contains nests of chondroblasts whose configuration and programming allow circumferential growth of the physis. Within the zone of Ranvier, deep to the chondroblasts, are osteoblasts that provide bony attachment to the perichondral ring. Surrounding the zone of Ranvier, and continuous with it, the tissue type changes to predominantly fibroblasts which form a tough fibrous ring known as the perichondral ring of Lacroix. This supporting structure joins the periosteum surrounding the metaphysis to the epiphyseal perichondrium and joint capsule–ligament complex. Interestingly the periosteum is only loosely applied in its attachment to the metaphysis at this level allowing a certain amount of tensile displacement of the epiphysis and physis before tightening as a checkrein to prevent further damage. It is thought that this may be a mechanism through which overgrowth occurs after physeal injury. Longitudinal growth stimulation secondary to joint loading forces is usually tempered by the intrinsic tensile restraint imposed by the cylindrical periosteal sleeve. When this is temporarily released as in a fracture, overgrowth may occur.15 Further structural support is provided by the formation of a lappet. A lappet is broadly defined as a fold of tissue, and in this situation refers to the fold of tissue encompassing a portion of the peripheral physis and the zone of Ranvier that overhangs and attaches to the metaphysis. Its job is to assist in preventing damage to the physis by resisting the action of shear forces. Physeal injuries have been classified by many authors, however the classification of Salter and Harris16 has stood the test of time due to it being easy to remember and having clinical application in guiding treatment and identifying injuries which may be prone to growth disturbance (Figure 4). Progressing through the categories, from I to V, one generally sees an increasing incidence of growth arrest, although it must be borne in mind that growth arrest is a multifactorial issue in which other anatomic and mechanistic features will play a part in determining the outcome. A type I injury is a pure physeal separation. The physis remains attached to the epiphysis. Aside from this occurring at the proximal femur or radius where there are vascular issues, this has a low risk of growth disturbance. A type II injury has a transverse physeal fracture line running predominantly through the hypertrophic zone and exiting through the metaphysis. This attached piece of metaphysis is known as the Thurston-Holland fragment. The germinal zone of the physis remains with the epiphysis, is usually undamaged and the risk of growth disturbance is low. A type III injury is the first of the articular fractures with the fracture line running transversely across the physis through the hypertrophic zone then deviating through layers of the physis, including the germinal zone, through the epiphysis into the joint. A type IV injury has a vertically orientated fracture line passing through metaphysis, physis, including the germinal zone, and epiphysis into the joint. Both type III and IV injuries require an anatomic reduction to prevent intra-articular joint deformity and to align the zones of the physis to help prevent the formation of a bony bridge and subsequent growth disturbance. A type V injury is a crush injury, affecting all layers of the physis and constitutes a very high risk of growth disturbance. Various authors have added extra categories to this classification to include, for example, bone loss, however, these are very rare fracture types and perhaps overcomplicate what is, in its basic form, a useful classification. There are three main factors determining the likelihood of growth disturbance following physeal injury. One is structural injury to the cells of the germinal zone, the second is an injury to the blood supply of the physis and the third is the formation of a bony bridge across the injured physis, acting as a tether. Whether any growth disturbance is clinically relevant will have more to do with the anatomic location of the fracture, any residual deformity, the age of the patient and the amount of growth remaining. With regards to structural injury to the germinal zone, the morphology of the physis, the energy of injury and the classification of the type of physeal injury (previously discussed) are all important factors. Not all physes conform to the textbook disc shape with uniform structure of zones as described a few paragraphs earlier. Some physes are rather undulating which is thought to be partly in response to the forces acting across it, as the delicate layers of cells within the physis align themselves perpendicularly to those forces to minimize the effect of shear, to which the physis is more vulnerable. The distal femoral physis, for example, has an undulating appearance as can be appreciated on standard AP and lateral radiographs. Any fracture line passing transversely through this particular physis is more likely to damage different zones and is therefore more likely to injure the biologically important zones such as the germinal zone. The distal radial physis is, on the other hand, much more flat and disc-like, hence the commonly seen Salter–Harris type II injuries have a fracture line that passes much more uniformly through the hypertrophic zone keeping away from the zones that may cause growth disturbance. The distal radius and distal femur are also good contrasting examples to illustrate the importance of the energy of the injury to growth arrest. Distal radial fractures in children are often low energy injuries following a fall from standing height. The low energy characteristic will tend to limit injuries to structures in the immediate vicinity of the fracture line, hence sparing the germinal zone of the physis in most cases. A distal femoral physeal injury, on the other hand, is not a low energy injury and is often observed with a high-energy mechanism such as falling from a tree. The zone of injury is much more extensive and subsequently the chance of damage to the germinal zone of the physis is increased. Both the structural differences between the distal femoral and distal radial physes, and the likely differences in energy of injury, explain why the risk of growth arrest after apparently similar Salter–Harris type II injuries is rare after a distal radial fracture and significantly more common after a distal femoral injury. Some anatomic regions also seem to be predisposed to certain physeal injury patterns. Most physeal injuries of the distal radius, radial neck and proximal humerus are of a Salter–Harris type II injury. The distal tibia and ankle show more of a mixed pattern, regularly presenting with a range of different physeal injuries of Salter–Harris types II, III, IV. With regards to the blood supply of the physis, there are two distinct types that were identified by Dale and Harris17 in their set of classic experiments. Type A epiphyses are nearly completely covered in articular cartilage. Blood supply enters the physis by traversing the perichondrium and is therefore vulnerable to injury should there be a separation of the epiphysis from the metaphysis. Much more robust are the type B epiphyses where the blood supply enters from the epiphyseal side, and in the event of a separation of the epiphysis from the metaphysis, the blood supply to the germinal layer of the physis is more likely to be preserved. In the human, there are only two type A epiphyses, those of the proximal femur and the proximal radius. The epiphysis and the sleeve fracture The long bone epiphyses are purely cartilaginous at birth except for that of the distal femur. Throughout growth these cartilaginous epiphyses gradually develop secondary ossification centres, which appear centrally and expand centrifugally to take over almost the entire end of the long bone. The chondroepiphysis is strengthened by the periosteum, which blends with its outer surface. As the secondary ossification centre develops it creates an area of increased stiffness inside the more flexible cartilaginous periphery. This predisposes to certain fracture patterns, usually incorporating a portion of the ossification centre itself. Purely cartilaginous epiphyseal fractures can occur and may be of the shell or sleeve type such as a patellar fracture or tibial spine avulsion. These are generally identifiable on a radiograph, but many are more difficult to identify and MRI may be required for diagnosis. Plastic deformation Plastic deformation is commonly seen in the spectrum of paediatric long bone injury. The environment in which this injury occurs is very specific and is much more likely to occur in children's bones as they are weaker in bending strength and absorb more energy before fracture. As force is applied to the immature bone elastic deformation occurs, which by definition, when released, allows the bone to revert to its original shape. Should more force be applied then a series of microscopic failures of the bony architecture can occur, causing deformation, but upon release leaves the bone with a residual deformity. This is plastic deformation. If more force is applied it will eventually result in the complete macroscopic failure (or fracture) of the bone. Unless one suspects plastic deformation however, it is easy to miss, for example, as the underlying cause of a “Monteggia variant” injury in the forearm where the radial head is dislocated, but no classic ulna fracture is identified (Figure 5). On closer inspection, the usual straight border of the ulnar diaphysis on the lateral radiograph is lost, and has a bowed appearance, its apex in the direction of the dislocation. Only the identification of the plastic deformation of the ulna and its subsequent correction will allow spontaneous reduction of the radial head for a satisfactory outcome. Biological aspects of fracture healing  The well-described stages of fracture healing in the child are very similar to those in the adult. An inflammatory phase consisting of haematoma formation, local necrosis, cellular proliferation, haematoma organization and endosteal repair is followed by the reparative phase consisting of cellular organization and finishes with the remodelling phase. There are notable differences with this process in the child that may account for the ability of a child's fracture to retain some inherent stability, heal more quickly and remodel more effectively than the adult and these will be discussed here. Healing of the fracture in the diaphysis or metaphysis is more analogous to the adult system, in contrast many differences occur in healing of physeal and epiphyseal injuries, which under ideal circumstances, are programmed to occur without the classic callus response. The advantage of this callus-free system is that it may restore growth potential to these specialized tissues. Should the more “adult” system of callus-based healing occur then the likelihood of osseous bar formation, growth disturbance and deformity may increase after physeal or epiphyseal injury.15 Periosteal activation In a metaphyseal or diaphyseal fracture the haematoma phase differs from that of the adult, in that the periosteum peels away from the bone much more readily and extensively after trauma, allowing significant haematoma formation from the high concentration of blood vessels in this area. This extensive haematoma formation is a real driving force to subsequent phases in healing with callus. Membranous ossification occurs beneath the periosteum in diaphyseal and metaphyseal fractures and supplements endochondral ossification occurring at the bridging of the fracture gap. The contribution of membranous ossification is greater in the child compared to the adult due to this highly vascular and reactive periosteum, which is loaded with osteoblasts. As has been described earlier, the periosteum is much more adherent at the end of the bone around the physis, and is continuous with the perichondrium of the epiphysis. This minimizes periosteal detachment in epiphyseal and physeal trauma and thus limits the ability of these special injuries to heal with undesirable callus. Vascular response One normally sees the early vascular response on the radiograph in the first few weeks of fracture healing. This is manifested by an increased lucency around the fracture site secondary to a hyperaemic response and local necrosis of the bone ends. Due to the more impressive vascular system within and surrounding the bone in the child this process is more potent and visually apparent than in the adult. Vascularity is key in many aspects of fracture healing including the conversion of cartilaginous callus to bone. Lack of appropriate blood supply can prevent this stage of the fracture healing process and result in a non-union. Due to the excellent vascular supply in children this situation is unusual, thus the incidence of non-union is lower than that of an adult. Structural composition Mature cortical bone is slow to heal. The relatively high ratio of vascular trabecular bone to mature cortical bone in the metaphysis in a child can help to explain the rapidity in which metaphyseal fractures heal, when compared with an adult, whose metaphysis consists of a higher ratio of mature cortical bone. Remodelling In contrast to fractures in the adult, those in the child have an amazing capacity to remodel to improve the appearance and function of a deformed limb. Deformities which would be considered unacceptable as an end point for fracture treatment in the adult may, in fact, be entirely satisfactory in the child in the knowledge that with the benefit of Wolff's law and enough growth remaining the same deformity can be completely remodelled. Wolff's law simply states that a bone remodels according to the stresses placed across it. The criteria for an acceptable reduction are based on the anticipated remodelling. The process of remodelling is based upon the coordinated actions of osteoclasts and osteoblasts. Bone is added to the concavity of the deformity whilst bone is removed from the convexity of the deformity, thus gradually correcting the bone towards its optimal anatomic shape. Remodelling does not just occur at the fracture site itself. Remodelling also occurs significantly at the physes at either end of the long bone. This is why one may find that even with a residual anatomic deformity of a femoral diaphyseal fracture, the mechanical axis of the entire limb may well have returned to normal (or near-normal) due to the contribution of the physes to remodelling. Axial shortening and angular deformity tend to remodel well, whilst rotational deformity has little remodelling potential. Whilst a child's bone has an ability to remodel, one needs to know the limits of what is achievable in each particular injury so that realistic outcomes can be expected. There are three main factors that determine the remodelling potential of a fracture. The first and most important is the child's age: the younger a child is, the more remodelling capacity is available due to the time available left for growth. With growth, the bone increases in diameter as well as length. The remodelling malunion can be effectively “swallowed up” by the growing bone, engulfing a region of remodelling thus using normal growth to its advantage. The second factor is the proximity of the fracture to a physis, and in particular a fast growing physis. The humerus gains 80% of its growth from the proximal physis at the shoulder and only 20% from the distal physes at the elbow. In the forearm, 75% of the growth comes from the distal radius and 25% from the proximal radius. It therefore stands to reason that proximal humeral and distal radial fractures have significantly more remodelling potential than the commonly seen distal humeral supracondylar fracture or proximal radial diaphyseal fracture. The diaphysis of a long bone does not tend to remodel well in its own right. The appearance of callus and new bone formation after a diaphyseal fracture can make the deformity look less angular but true remodelling is more limited. This is in contrast to the fractures positioned close to the ends of the bone. The third factor determining remodelling potential is the plane of the fracture. Remodelling occurs best when the plane of the fracture occurs in a plane of motion of the joint near which the fracture is located: the more movement in that plane, the larger the remodelling potential. Distal radial injuries that initially heal with apex volar angulation thus remodel more satisfactorily than do those with apex radial or ulnar angulation. The reason for this phenomenon can be attributed to Wolff's law. One should also consider that joints which have a large anatomical range of movement in a particular plane, tend to have a proportionally larger functional range of movement in the same plane, leaving some “spare” to account for any residual deformity that remodelling does not take care of. When one puts all these factors together it is apparent that the remodelling potential of a distal radius fracture with apex volar angulation in a 3-year-old child is tremendous and has markedly different characteristics to that of a 6-year old with a distal humeral supracondylar fracture left with a varus malunion, who will be unlikely to show significant remodelling. Alongside the biological and mechanical features described above that contribute to remodelling potential, this multifactorial process may also include other factors related to the environment. Children are very active and therefore tend to continually stress their skeletons. Bone, being such a dynamic tissue, will tend to respond to the forces applied to it in a favourable manner, not only to assist with growth and development, but to direct any remodelling that may be required. Growth arrest lines A radiographic characteristic worthy of discussion is that of growth arrest lines (Harris/Park lines) (Figure 6). These transverse trabecular striations run parallel to the physeal contour and are visible after periods of trauma or generalized illness, and represent temporary cessation of longitudinal bony growth. Because of the slowdown in growth, the trabeculae assume a transverse orientation forming an osseous plate as they get backed up rather than the longitudinal orientation that they normally assume. When longitudinal growth resumes, the discrete plate can then be seen on the radiograph over time as it increases in distance from the physis. The transverse lines in corresponding limbs can be used to measure growth resumption and limb-length discrepancy after a long bone fracture. In the event of partial physeal growth arrest, the line will appear oblique rather than parallel to the physis often giving the first clue that a deformity is developing. Growth stimulation Another characteristic of the healing of children's long bone fractures is that of growth stimulation of the injured bone and limb. A good example of this characteristic is in that of the femoral fracture. This phenomenon has presumably evolved to equalize limb lengths, as biologically, the child's body is expecting to unite a femoral fracture with a certain amount of shortening. The child's femur has the capacity to correct up to 2 cm of residual shortening. The mechanism for this process is unclear and not simply due to a hyperaemic response, as evidenced by the fact that fractures of the radius do not have this propensity to overgrow, and that anatomic reduction of femoral shaft fractures does not always result in the amount of overgrowth that is expected.8 Care of the multiply injured child  Management of the multiply injured patient can be daunting, even more so when the patient happens to be a child. Any orthopaedic surgeon treating children needs to be comfortable with the emergency management of multi-trauma in this patient group. Despite the apparent initial complexities of the multiply injured patient, the emergency management of life threatening trauma has been revolutionized by the Advanced Trauma Life Support (ATLS) system. It is in some way reassuring that treating the multiply injured child follows the same principles as those of an adult, however there are differences that the physician needs to be aware of. A: airway and cervical spine control The young child has a proportionally bigger head than the adult. This persists until 6 years of age. When transferring the younger injured child, cervical spine immobilization is critical, but a regular adult style spine board will tend to flex the head and thus the cervical spine, which is undesirable in the presence of suspected spinal injury. To achieve neutral spinal immobilization, a spine board with a cut away section for the occiput is necessary. Should this specialized equipment be unavailable then the child's thorax can be slightly raised using a layer of towels, or a roll, on an adult spine board to keep the spine in neutral alignment (Figure 7). With the cervical spine controlled the airway may be maintained by chin lift or jaw thrust. Being obligate nose breathers it is important to keep an infant's nostrils clear of mucus, vomit or foreign body. A child's airway differs anatomically from that of an adult. A child has a relatively larger tongue in relation to the size of the oral cavity and a shorter, more flexible epiglottis than that of an adult, which makes visualization of the vocal cords and intubation more challenging. A child also has a higher volume of lymphoid tissue in the nasopharynx which can contribute to airway obstruction. B: breathing Hypoxia is noted in up to 50% of comatose children with head injury, which in combination with hypotension at initial presentation has an 85% mortality.18 The need for early control of airway patency with special consideration of the anatomical peculiarities of children, for oxygen supplementation and endotracheal intubation for definitive control of the airway is not questionable. Endotracheal intubation is more difficult in the child and the literature describes several series of life threatening complications in association with attempted and failed intubations, particularly at-scene, by less experienced operators. A need is therefore highlighted for paediatric airway management to be carried out by personnel with specialist training and expertise. Ventilation should be confirmed by auscultating breath sounds in both lung fields. Abnormality may be due to pneumothorax, however, due to the shorter trachea of a child's airway in relation to the rest of their body it is easy to intubate the right main bronchus necessitating one to consider an improperly placed airway as the cause. C: circulation Physiologically the child compensates very well in comparison to an adult. The downside to this resilience is that shock is more difficult to identify. Hypovolaemia in the child is characterized by vasoconstriction of the arterioles, which maintains the blood pressure and may allow compensation for a loss of 25% of circulating volume with minimal alterations in vital signs. By the time a child becomes hypotensive, they have little more in reserve. It is a sign of significant concern. Tachycardia is usually the first measurable response to shock in children, but can also be stimulated by pain and anxiety complicating its use as a hard sign of physiological compromise. As shock progresses, the systolic blood pressure may fall, but the pulse pressure narrows, maintaining a normal mean arterial blood pressure. In addition, a large ratio of surface area to body weight in a child contributes to the rapid development of hypothermia, which can cause or exacerbate metabolic acidosis, induce pulmonary hypertension and increase hypoxia.19 Hypotension is likely to be caused by internal bleeding, as blunt trauma is the most common mechanism of injury. Internal bleeding is most likely to arise from abdominal or thoracic injury. Hypotension is rarely caused by pelvic fracture, isolated long bone fracture or closed head injury in the child. Hypotension must be dealt with immediately. There are numerous methods of securing vascular access in the child, many the same as the adult, but one can also include umbilical artery cannulation and access from the scalp veins in neonates and small babies as options. Intraosseous cannulation is also a useful technique in younger children. Whilst other forms of vascular access in the hypotensive child can be demanding even in experienced hands, the intraosseous route is simple and effective. Large volumes of fluid can be given in this way when combined with a pressure infusion device. Bones that can be used for this modality include the tibia, femur, iliac crest, clavicle, sternum and calcaneum. Unfortunately the sampling of blood from the intraosseous access has not been shown to accurately reflect haemoglobin or electrolyte levels in central venous blood.20 No adverse effects of intraosseous infusions have been identified with respect to infection, or growth of the bone, however, there have been a few reports of compartment syndrome after prolonged, rapid infusion. When replacing fluid, one must be mindful of the normal circulating volume of the child and the likely volume of blood loss. A child's circulating blood volume can be estimated as 80 mL/kg. A child's weight in kilograms can be estimated as weight (kg) = (age [yr] × 2) + 8. Fluid resuscitation begins with a crystalloid bolus equal to 25% of the circulating blood volume, which equates to approximately 20 mL/kg. If tachycardia or other signs of hypovolaemia persist after two crystalloid boluses, consideration should be given to transfusion of packed red blood cells.8 A small volume of blood loss, ordinarily of little concern in the adult, may constitute a significant haemorrhage in the child. Tolerances are tighter in the child due to the smaller fluid volumes and it is easy to create significant internal fluid shifts with excessive fluid replacement leading to lower arterial oxygenation from interstitial pulmonary oedema, particularly in the face of pulmonary contusion.21 Accurate input and output monitoring are essential often requiring central venous access and a urinary catheter. Normal urine output in an infant is 1–2 mL/kg/h and in a child or adolescent 0.5 mL/kg/h. Secondary survey The secondary survey should be completed as with any multiply injured patient to identify other injuries that may need treatment. The Glasgow Coma Scale (GCS) should be recorded. The original GCS, as used in adults, is difficult to apply in young children, in particular the verbal component of the scale. The British Paediatric Neurology Association (BPNA) has published a child's GCS (2001), based on the work of their GCS audit group, and recommends a split in how the verbal and motor components are scored at the age of 5 years.22 Other scoring systems are available, but on the basis that most physicians are familiar with the GCS in the adult setting, it seems sensible to use its modification when assessing a child in an emergency situation (Table 1). The trauma series of chest, lateral cervical spine and pelvis radiographs should be ordered, in series with the primary survey. Upon secondary survey, consideration should be given to obtaining standard two plane radiographs of any extremity suspected of having sustained significant trauma as evidenced by tenderness, swelling, abrasion or ecchymosis. In the case of head trauma a CT scan is valuable for the assessment of potential intracranial injury. Modern CT scanners can provide detailed imaging, in such a rapid fashion, that should thoracic, abdominal or pelvic injury initially be suspected, then these anatomic regions should be included at the same sitting rather than having a severely injured child make a repeat trip to the scan room. This obviously necessitates the coordinated multidisciplinary approach of all the surgical teams to ensure that everybody's diagnostic wishes are catered for, whilst also being mindful that the CT scanner is no place for an unstable critically ill patient. Resuscitation always comes first. The routine use of comprehensive blood screening in paediatric trauma is controversial. Samples are often difficult to obtain, expensive to process and in the majority of cases are non-contributory. A selective policy of blood testing using clinical judgement is most appropriate in the severely injured child, with the useful tests being a full blood count, urea and electrolytes, glucose, blood gas analysis and, in the case of head injury with a reduced GCS, a coagulation screen. Coagulopathy, because of a release of brain tissue plasminogen activator, is well documented after traumatic brain injury and may lead to secondary brain injury if left untreated.23 Blood gas analysis is also useful in severely injured children, with an admission base deficit of less than −5 being a meaningful predictor of injury severity and mortality in the child. The failure to clear this base deficit is associated with a mortality rate of up to 100%.19 It has also been described that admission base deficit is more sensitive than vital signs in detecting shock in the child. Blood typing and cross-match may be required depending on the severity of injury to the child and likelihood of needing a blood transfusion. Open fracture management Open fractures in children are significant injuries, however the characteristics, management and potential outcome differ in many ways from the adult. The overall infection rate of a paediatric open fracture is 3%. When classified by Gustilo and Anderson grade, type I fractures have an incidence of infection of 2%, type II fractures an incidence of 2%, and type III fractures an incidence of 8%.24 These are lower than the accepted figures of infection following open fractures in the adult population (Table 2).  | Type I | Wound less than 1 cm long |  |  | Clean |  |  | Little soft tissue damage |  |  | Type II | Wound more than 1 cm long |  |  | No extensive soft tissue damage, flaps or avulsions |  |  | Slight or moderate crush injury |  |  | Type III | A | Any high-energy trauma regardless of size of wound |  |  | Adequate soft tissue coverage despite extensive laceration, flaps or avulsion |  |  | B | Extensive soft tissue injury with periosteal stripping |  |  | Requires local or free flap coverage |  |  | Usually associated with massive contamination |  |  | C | Associated with arterial injury requiring repair |  | | | |
The use of antibiotics is important in the management of open fractures and the timing of their administration appears to be crucial. Wilkins and Patzakis reported an infection rate of 4.7% in 1104 open fractures in both adults and children when antibiotics were administered within 3 h of the injury compared with 7.4% when more than 3 h had elapsed since injury.25 Furthermore antibiotic administration is the most important factor associated with treatment in reducing the infection rate. Standard recommendations for antibiotic prophylaxis in children are in line with those for adults. A first generation cephalosporin given intravenously is an ideal choice for administration to all children with an open fracture, except those with known allergy to penicillin or cephalosporins. In these children clindamycin is commonly used. In severe type II or type III injuries, gentamicin is added to the regimen for wider gram-negative cover, and one may also add penicillin to cover clostridium species and other anaerobes in patients with farm or vascular injuries.26 There is no evidence for the use of second tier antibiotics in the initial management of open fractures in children, except for patients with an allergy to cephalosporins. The duration of antibiotics is also a matter of debate, with most authors using a minimum of 48 h of intravenous administration. Short-term use of between 48 and 72 h appears to be as effective as long-term use with longer durations of administration having little effect on the overall infection rate. Standard teaching recommends the emergency operative treatment of open fractures. The timing of surgery in the child with an open fracture may not be so critical as once thought. In a multi-centre series of 554 open fractures in children there was no significant difference in the infection rate between the fractures that had been operated on within 6 h of the injury and those that had been treated at least 7 h after the injury. The authors concluded that irrigation and debridement of open fractures in children can be performed within the first 24 h after injury without increasing the risk of infection as long as intravenous antibiotics are started on presentation to the emergency department.24 Other authors support these findings and whilst this is an acceptable management principle in the majority of open fractures, one must judge each case on its own merits. All open fractures are not the same and should an open fracture be associated with neurovascular compromise, severe soft tissue injury, or other urgent operative indication then emergency irrigation and debridement are recommended. This quite rightly places the emphasis on accurate and thorough clinical assessment of the patient to determine the most appropriate surgical plan. All open fractures in children require formal irrigation and debridement. This includes type I injuries, despite often having an innocuous looking puncture wound. It is easy to see how merely squirting irrigation fluid into such a puncture wound has little chance of its intended purpose to flush out foreign and contaminated debris effectively. Whilst there appears to be evidence to the contrary in the literature, one must be mindful that these are small series with limited power. The principles of initial management are the same as those in an adult of extending the wound edges to obtain a clear visualization of the injured tissue, to perform thorough irrigation, to manually excise foreign material or any grossly devitalized tissue, clean the bone ends, perform fasciotomies where appropriate and stabilize the fracture. The difference in children occurs with the extent of the debridement. Due to the excellent vascular supply in the child and remarkable healing potential of their soft tissue, it is recommended that only grossly devitalized tissue is removed. Soft tissue of dubious viability that would ordinarily be excised in the adult may be left and reassessed at the second-look surgery. Bone that is stripped of its soft tissue attachments which requires removal in the adult may often be left in the child. It will usually incorporate and bone defects reconstitute in the presence of an intact periosteal sleeve. Type I injuries can usually be closed primarily after the first irrigation and debridement. Type II and III injuries will require a return to the operating room for further debridement each 48–72 h until the tissues look clean and healthy. Irrigation should be performed with a low-pressure technique using soap or detergent solution, which is shown to be more effective than saline solution alone or antibiotic irrigation. Routine wound cultures are not indicated in the primary management of open fractures due to poor correlation with the organisms that ultimately cause infection in individual cases. Wound cultures are only of use once infection has developed. For further detail and the specific management of open fractures in children according to their anatomic region the reader is directed to an excellent review on the subject by Stewart et al.26 Outcome of paediatric trauma  Mortality after trauma is a useful, easily defined parameter, but it does not accurately reflect the significant burden of injury on those who survive. Trauma is the leading cause of preventable long-term disability in children. To comprehensively understand the outcome of trauma one needs some kind of measurement tool with which to work, one that is reliable, valid and responsive. Measuring the outcome of trauma in an adult population is known to be difficult. Let us then consider the requirements of a tool to measure the outcomes of trauma in children. Marked differences exist between how children and adults report their health and these need to be considered when designing or choosing an outcome instrument. Other factors that need to be incorporated are that children have different social roles involving family, peer group, school and their community. The younger the child the less able they may be to actually report their health status. It is not until a child reaches 5 years of age that they are considered reliable sources of their own health status. Age appropriate language will need to be used. Parents may need to fill out a questionnaire on behalf of the child, which requires an agreement between the child and their parent on the health status, already known to be inaccurate. Even the ability of paediatric patients to maintain attention to fill out a lengthy questionnaire will need to be considered. Any instrument measuring trauma outcomes needs to be able to assess patient changes over time and importantly needs to take into account normal growth and development. Many childhood outcome measures have been reported in the literature, but none have been rigorously validated for use in the paediatric trauma population.27 Bearing this in mind, there are a number of trends and issues seen in the outcome of trauma in children that warrant discussion. As a rule of thumb in children, the worst injury sustained will predict the mortality and long-term morbidity, which is not always the case in adults. Children sustaining multiple trauma with a head injury tend to have worse outcomes than those with no head injury; however, the recovery from head injury in childhood is substantially better than in adults. Even children who are in a coma for several days may make a full motor recovery, although these children may show cognitive deficits and psychological sequelae. One should, therefore, assume a full neurological recovery, and thus provide full and complete orthopaedic care. It is inappropriate to leave a musculoskeletal injury untreated while following the recovery of a head injury in a child.21 Children with moderate to severe injuries and no head injury still demonstrate notable disability at discharge, with fractures contributing significantly to the level of disability. It is encouraging that children are known to make a more rapid musculoskeletal recovery than do adults, but attaining normal function as compared to the function of their peers takes a considerable amount of time. The highest level of physical disability is seen in the first month following injury. By six months children have usually still not reached the same level of functioning as their peers. This trend is particularly seen when lower extremity injury is present. Although true long-term data are unavailable for children, it has been shown that any disabilities that still exist in children one year post-injury are long term and possibly permanent.28 Persistent dependence in mobility or self-care functioning in children after trauma can disrupt both the educational activities of the children and the employment of their parents, resulting in significant individual and societal costs. Up to 40% of families have been shown to report some problem with finances or parental work several months after injury.29 Psychological aspects  Although the physical injury may be the most obvious, the injury-related psychosocial disturbances may be long-lasting. 30% of traumatized children without head injury, as well as 48% of those children who had suffered head injury, displayed evidence of an emotional or behavioural disturbance for an average of 2.5 years after the trauma.30 Family members of injured children are also prone to psychological dysfunction. One study found an incidence of emotional changes in 46% of uninjured siblings, and that 40% of parents reported changes in their marital relationship after a child had been injured. Single-parent families show a much higher risk of disruption of family than conventional two-parent families.31 Any physician treating an injured child must bear in mind the psychological aspects of the injury on the child and family as well as the apparently more pressing requirement to fix a broken bone. Consideration of this aspect of care will enhance the doctor–patient–family relationship, which will ultimately be of benefit to the injured child. Summary  The response of children to trauma is different to that of the adult. Knowledge of how the child responds anatomically, physiologically and psychologically to trauma is of paramount importance when treating the injured child. A sound knowledge of the scientific principles underlying these responses will help the orthopaedic surgeon provide the most appropriate care with the best chance of a satisfactory outcome. References  1. 1Pearson J, Stone DH. Pattern of injury mortality by age-group in children aged 0–14 years in Scotland, 2002–2006, and its implications for prevention. BMC Pediatr. 2009;9:26.
CrossRef
2. 2Guyer B, Gallagher SS. An approach to the epidemiology of childhood injuries. Pediatr Clin North Am. 1985;32:5–15. MEDLINE 3. 3Spady DW, Saunders DL, Schopflocher DP, Svenson LW. Patterns of injury in children: a population-based approach. Pediatrics. 2004;113:522–529. 4. 4Danseco ER, Miller TR, Spicer RS. Incidence and costs of 1987–1994 childhood injuries: demographic breakdowns. Pediatrics. 2000;105:E27. 5. 5Wilkins K. The incidence of fractures in children. In: Rockwood C, Wilkins K, Beaty J editor. Fractures in children. 4th edn. Philadelphia: Lippincott-Raven; 1996;p. 3–17. 6. 6Chambers RB. Orthopaedic injuries in athletes (ages 6 to 17). Comparison of injuries occurring in six sports. Am J Sports Med. 1979;7:195–197. MEDLINE |
CrossRef
7. 7Purcell L, Philpott J. Trampolines at home and playgrounds: a joint statement with the Canadian paediatric society. Clin J Sport Med. 2007;17:389–392.
CrossRef
8. 8Herring J. General principles of managing orthopaedic injuries. In: Herring J editors. 4th edn. Tachdjian's pediatric orthopaedics. Vol. 3:Philadelphia: Saunders Elsevier; 2008;p. 2335–2389. 9. 9Rennie L, Court-Brown CM, Mok JY, Beattie TF. The epidemiology of fractures in children. Injury. 2007;38:913–922. Abstract | Full Text |
Full-Text PDF (190 KB)
|
CrossRef
10. 10Banerjee S, Barry MJ, Paterson JM. Paediatric pelvic fractures: 10 years experience in a trauma centre. Injury. 2009;40:410–413. Abstract | Full Text |
Full-Text PDF (186 KB)
|
CrossRef
11. 11Silber JS, Flynn JM, Koffler KM, Dormans JP, Drummond DS. Analysis of the cause, classification, and associated injuries of 166 consecutive pediatric pelvic fractures. J Pediatr Orthop. 2001;21:446–450. MEDLINE |
CrossRef
12. 12Leonard M, Sproule J, McCormack D. Paediatric spinal trauma and associated injuries. Injury. 2007;38:188–193. Abstract | Full Text |
Full-Text PDF (161 KB)
|
CrossRef
13. 13Martin BW, Dykes E, Lecky FE. Patterns and risks in spinal trauma. Arch Dis Child. 2004;89:860–865.
CrossRef
14. 14Bilston LE, Brown J. Pediatric spinal injury type and severity are age and mechanism dependent. Spine (Phila Pa 1976). 2007;32:2339–2347. 15. 15Ogden J, Ganey T, Ogden D. The biological aspects of children's fractures. In: Rockwood C, Wilkins K, Beaty J editor. Fractures in children. 4th edn. Philadelphia: Lippincott-Raven; 1996;p. 19–52. 16. 16Salter R, Harris W. Injuries involving the epiphyseal plate. J Bone Joint Surg. 1963;45:587–622. 17. 17Dale G, Harris W. Prognosis of epiphyseal separation: an experimental study. J Bone Joint Surg. 1958;40:116. 18. 18Meyer PG. Paediatric trauma and resuscitation. Curr Opin Anaesthesiol. 1998;11:285–288. MEDLINE |
CrossRef
19. 19Randolph LC, Takacs M, Davis KA. Resuscitation in the pediatric trauma population: admission base deficit remains an important prognostic indicator. J Trauma. 2002;53:838–842. MEDLINE 20. 20Fairgrieve R, Lerman J. Anaesthetic care of the injured child. Curr Opin Anaesthesiol. 2000;13:141–145. MEDLINE |
CrossRef
21. 21Tolo V. Management of the multiply injured child. In: Beatty JH, Kasser JR editor. Fractures in children. 4th edn. Philadelphia: Lippincott-Raven; 1996;p. 83–85. 22. 22BPNA. Child's Glasgow Coma Scale. 2001. http://www.bpna.org.uk/audit/GCS.PDF. 23. 23Keller MS, Coln CE, Trimble JA, Green MC, Weber TR. The utility of routine trauma laboratories in pediatric trauma resuscitations. Am J Surg. 2004;188:671–678. 24. 24Skaggs DL, Friend L, Alman B, et al. The effect of surgical delay on acute infection following 554 open fractures in children. J Bone Joint Surg Am. 2005;87:8–12. MEDLINE 25. 25Wilkins J, Patzakis M. Choice and duration of antibiotics in open fractures. Orthop Clin North Am. 1991;22:433–437. MEDLINE 26. 26Stewart DG, Kay RM, Skaggs DL. Open fractures in children. Principles of evaluation and management. J Bone Joint Surg Am. 2005;87:2784–2798. MEDLINE 27. 27Willis CD, Gabbe BJ, Butt W, Cameron PA. Assessing outcomes in paediatric trauma populations. Injury. 2006;37:1185–1196. Abstract | Full Text |
Full-Text PDF (161 KB)
|
CrossRef
28. 28Winthrop AL, Brasel KJ, Stahovic L, Paulson J, Schneeberger B, Kuhn EM. Quality of life and functional outcome after pediatric trauma. J Trauma. 2005;58:468–473. MEDLINE 29. 29Aitken ME, Jaffe KM, DiScala C, Rivara FP. Functional outcome in children with multiple trauma without significant head injury. Arch Phys Med Rehabil. 1999;80:889–895. Abstract |
Full-Text PDF (654 KB)
|
CrossRef
30. 30Basson MD, Guinn JE, McElligott J, Vitale R, Brown W, Fielding LP. Behavioral disturbances in children after trauma. J Trauma. 1991;31:1363–1368. MEDLINE 31. 31Dealy M, Warman J. Psychological aspects of the injured child. In: Rockwood C, Wilkins K, Beaty J editor. Fractures in children. 4th edn. Philadelphia: Lippincott-Raven; 1996;p. 97–101. 32. 32Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am. 1976;58:453–458. MEDLINE 33. 33Gustilo RB, Mendoza RM, Williams DN. Problems in the management of type III (severe) open fractures: a new classification of type III open fractures. J Trauma. 1984;24:742–746. MEDLINE Simon P Kelley MBChB FRCS (Tr and Orth) Trans-Canada Paediatric Orthopaedic Fellow at the BC Children's Hospital, 4480 Oak Street, Vancouver, British Columbia V6H 3V4, Canada PII: S1877-1327(09)00192-4 doi:10.1016/j.mporth.2009.11.003 © 2009 Elsevier Ltd. All rights reserved. | |
|