Fractures are common during childhood; however, they can also be the presenting symptom of primary or secondary causes of bone fragility. The challenge is to identify those children who warrant further investigation. In children who present with multiple fractures that are not commonly associated with mild to moderate trauma or whose fracture count is greater than what is typically seen for their age, an initial evaluation, including history, physical examination, biochemistry, and spinal radiography, should be performed. In children with bone pain or evidence of more significant bone fragility, referral for specialist evaluation and consideration of pharmacologic treatment may be warranted.
Key points
- •
Fractures in children are common; 16% to 25% of children sustain more than 1 fracture by adulthood.
- •
In the absence of severe trauma, the presence of at least 1 vertebral compression fracture, 2 or more long bone fractures by 10 years of age, or 3 or more long bone fractures by 19 years is an indication for further bone health evaluation.
- •
Bone fragility in children can arise either from a primary bone disorder (such as osteogenesis imperfecta [OI]) or secondary to an underlying medical condition.
- •
Ensuring adequate calcium and vitamin D intake, weight-bearing exercise, and minimizing exposure to adverse bone factors are important interventions.
- •
In children with bone pain or significant bone fragility (vertebral compression fractures or bone mineral density [BMD] z score less than −2), referral for specialist evaluation is indicated.
Fractures are a common occurrence in the pediatric population. With a prevalence that is increasing over time, up to 25% to 40% of girls and 30% to 50% of boys sustain a single fracture by adulthood. Between 16% and 25% of children have more than 1 fracture. Multiple fractures can, however, be an indicator of underlying bone fragility. The dilemma, therefore, when evaluating an otherwise apparently healthy child with a history of multiple fractures, is to distinguish between the child with intact bone health and the child who warrants further detailed investigation.
In adults, BMD as measured by dual-energy x-ray absorptiometry (DXA) is used to help diagnose and define individuals with osteoporosis who have an increased future risk for fractures. In contrast, for children, the diagnosis of osteoporosis cannot be made using bone densitometric criteria alone. The 2013 guidelines of the International Society for Clinical Densitometry (ISCD) outlined that bone fragility in children should be defined by a clinical significant fracture history, with, but not mandating, a low BMD or bone mineral content (BMC) below a z score of −2 for age and gender. Understanding the determinants of fracture risk is important to identify children who clinically have a significant fracture history and, therefore, need further assessment.
The pattern of fractures in childhood
Fractures occur more commonly in boys, with a peak incidence at 11 to 12 years in girls and 13 to 14 years in boys, corresponding to periods of increased growth velocity. The rapidly growing adolescent skeleton undergoes a period of relative thinning of the bone cortices due to a delay between maximal longitudinal bone growth and peak bone mineral accrual ( Fig. 1 ).
The site of a fracture often depends on the mechanism of injury and the age of the child. Fractures of the distal forearm constitute the most common site of injury in all ages, accounting for 20% to 25% of fractures. Hand fractures, including the phalanges, carpals, and metacarpals, are the second most common fracture occurring in childhood but are seen usually as a result of crush injuries. In comparison, vertebral compression fractures are uncommon, accounting for only 1% to 5% of all fractures in childhood. The presence of vertebral body height loss should alert the clinician to the possibility of a bone fragility disorder.
The pattern of fractures in childhood
Fractures occur more commonly in boys, with a peak incidence at 11 to 12 years in girls and 13 to 14 years in boys, corresponding to periods of increased growth velocity. The rapidly growing adolescent skeleton undergoes a period of relative thinning of the bone cortices due to a delay between maximal longitudinal bone growth and peak bone mineral accrual ( Fig. 1 ).
The site of a fracture often depends on the mechanism of injury and the age of the child. Fractures of the distal forearm constitute the most common site of injury in all ages, accounting for 20% to 25% of fractures. Hand fractures, including the phalanges, carpals, and metacarpals, are the second most common fracture occurring in childhood but are seen usually as a result of crush injuries. In comparison, vertebral compression fractures are uncommon, accounting for only 1% to 5% of all fractures in childhood. The presence of vertebral body height loss should alert the clinician to the possibility of a bone fragility disorder.
Determinants of fracture risk
A child’s risk for having a fracture depends on 3 main factors: the severity of the trauma, the exposure to potential trauma, and the underlying bone strength, including bone density, size, and quality. In a child who undertakes regular vigorous activity and who is frequently exposed to episodes of moderate trauma, a history of multiple fractures may not be unexpected. In contrast, for a child who is exposed to infrequent minor episodes of trauma, fractures may be attributable to an underlying deficit in bone strength.
Severity of Trauma
Given enough force, any bone will break; however, the degree of trauma associated with a fracture relates to the underlying bone quality. Using high-resolution peripheral quantitative CT, children with fractures as a result of mild trauma have been shown to have reduced bone strength and cortical thickness compared with nonfracture controls. In comparison, the presence of a moderate trauma fracture is not associated with bone quality deficits. Thus, the presence of a low trauma fracture should alert clinicians to the potential of underlying bone fragility. This relationship seems to carry through to adulthood, with reduced bone strength, density, and cortical thickness seen in adult men and women with a history of mild but not moderate trauma fractures in childhood. Obtaining an accurate history of the mechanism and circumstances that led to a fracture is, therefore, important to ascertain whether the force associated with the injury seems plausible.
Potential Exposure to Trauma
Exercise and physical activity are associated with increased BMD. However, 50% of fractures in children are related to sporting activities. Higher bone mass associated with exercise does not seem to compensate for the increased injury exposure in children who participate in regular vigorous physical activity. Children who undertake daily vigorous physical activity have double the fracture risk of those who do fewer than 4 episodes per week, independent of their bone mass.
Bone Strength: Bone Density, Size, and Quality
Bone strength and its ability to resist fracture depend on bone density, size, and quality. BMD is clearly associated with future fracture risk in adults. Although in children this relationship is not as strong, lower BMD does seem to correlate with increased fracture risk. In a meta-analysis of 8 case-control studies, children with a recent fracture had a standardized mean BMD z score −0.32 lower than children without a history of fractures. A prospective study in children aged 9.9 years also demonstrated that for every 1 SD decrease in volumetric BMD there was an associated 12% increased risk for a subsequent fracture in the following 2 years. A normal BMD z score does not, however, preclude a diagnosis of underlying bone fragility. In a cohort of 66 children with low-trauma vertebral or multiple peripheral fractures, only 8% were found to have a BMD z score less than −2.
Genetic factors play a critical role in peak bone mass determination, with up to 80% of the variance in BMD attributable to heritability. Large genome-wide studies have identified more than 55 different loci associated with attainment of peak BMD. To date, however, these genomic regions and polymorphisms explain only 4% to 5 % of the phenotypic variation seen in adult BMD.
Bone strength is determined not only by BMD but also by bone size and quality. Although the evidence in pediatric cohorts is limited, in adults, nontraumatic vertebral compression fractures have been demonstrated to relate to bone size and trabecular bone microarchitecture independent of the bone density.
Conditions associated with bone fragility
When a child’s history of multiple fractures seems discordant with the severity of or exposure to trauma, further investigation is indicated to assess for potential underlying bone fragility.
Bone fragility can result either as a result of a primary bone condition or as a consequence of a secondary contributing factor or chronic condition ( Box 1 ). In all children, particularly infants, nonaccidental injury (NAI) should be excluded as a cause for multiple fractures. Although the pattern of fractures (such as posterior rib and metaphyseal fractures) and associated clinical signs of injury (such as retinal hemorrhage or bruises) may be contributory, differentiating NAI from OI in infants can be difficult.
Primary conditions
Impaired collagen gene expression or collagen post-translational modification
OI
Impaired collagen cross-link formation
Bruck syndrome
Connective tissue defects
Ehlers-Danlos syndrome
Marfan syndrome
Homocystinuria
Defective bone mineralization from low alkaline phosphatase activity
Hypophosphatasia
Impaired cell signaling and osteoblast function
Osteoporosis pseudoglioma syndrome
IJO (cause unknown)
Secondary conditions
Medication induced
Glucocorticoids
Antiepileptic medication
Anticoagulants
Reduced weight-bearing activity or muscle bulk
Duchenne muscular dystrophy
Cerebral palsy
Infiltrative conditions
Leukemia
Thalassemia
Chronic inflammatory conditions
Juvenile idiopathic arthritis
Inflammatory bowel disease
Endocrine abnormalities
Hypogonadism
Growth hormone deficiency
Hyperparathryoidism
Hypercortisolism
Vitamin and nutritional deficiencies
Vitamin D deficiency
Celiac disease
Anorexia nervosa
Cystic fibrosis
Renal disease
Chronic renal failure with secondary hyperparathyroidism
Idiopathic hypercalciuria
Primary Osteoporosis
OI is the most common genetic form of primary osteoporosis in children. The majority of cases are associated with mutations in 1 of 2 genes that encode the alpha chains of collagen type I (COL1A1 and COL1A2), although an increasing number of new genes involved in post-translational collagen modification have been described over the past 10 years. OI has a broad clinical phenotype, ranging in severity from perinatal lethality to mild clinical forms without fractures. Typical clinical findings, including blue sclera, gray opalescent primary teeth (dentinogenesis imperfecta), short stature, and joint laxity, can aid in the diagnosis. Radiographic features commonly include vertebral fractures, generalized osteopenia, and gracile long bones with evidence of bowing ( Fig. 2 ).
Children with connective tissue disorders, such as Ehlers-Danlos syndrome, can also present with joint laxity in addition to skin hyperelasticity, easy bruising, and recurrent joint dislocations. Connective tissue disorders can be associated with bone fragility, reduced BMD, and increased risk for fractures.
Idiopathic juvenile osteoporosis (IJO) is a condition that characteristically presents with bone pain and vertebral fractures prior to the onset of puberty. It is an important differential diagnosis to OI, particularly in the absence of typical OI clinical features or a positive family history. The underlying cause of IJO is unknown, but transiliac histomorphometry studies in patients with IJO demonstrate decreased trabecular bone volume, number, and thickness. Although many children with IJO seem to have spontaneous improvement in symptoms and BMD after the onset of puberty, in a subset of patients with more severe symptoms, ongoing bone pain, increased fracture risk, and vertebral deformities can persist into young adulthood.
Several rare genetic conditions involving defects in bone cell signaling and function or bone matrix homeostasis have also been implicated as causes of primary osteoporosis (see Box 1 ).
Secondary Osteoporosis
The presence of impaired bone health as a result of secondary chronic conditions is increasingly recognized in the pediatric population (see Box 1 ). Bone fragility can arise from a combination of factors, including the use of glucocorticoids, increased inflammatory cytokines, diminished nutrition, physical activity, and muscle bulk. Although in some children the secondary cause of the fractures may be clearly evident, a history of multiple fractures can be the presenting symptom for an otherwise clinically silent disease. Fractures and decreased BMD are not uncommon presentations for individuals with celiac disease. Acute lymphoblastic leukemia can present with fractures, even with a normal peripheral blood count. Secondary bone fragility can also occur early in the disease process, with 16% of children with acute lymphoblastic leukemia and 7% of children with a rheumatological condition having evidence of vertebral compression fractures within 30 days of diagnosis.
Nutritional deficiencies, in particular involving vitamin D, are an important contributor to bone fragility due to decreased bone mineralization. Vitamin D deficiency can arise from inadequate intake as a result of reduced sunlight exposure or decreased absorption from gut malabsorption, or from increased catabolism as seen with some of the antiepileptic medications.
Approach to the child with multiple fractures
Identifying the Child with a Clinically Significant Fracture History
The first step in the evaluation of an otherwise apparently healthy child who presents with multiple fractures is to assess whether the fracture history varies from the normal pattern of fractures seen in childhood and, as such, is clinically significant ( Fig. 3 ). Fractures as a result of severe trauma (falling from a height >3 m or as a result of a motor vehicle accident) are not unexpected and, therefore, should not contribute to a child’s fracture count. Similarly, crush fractures of fingers, toes, or nose are not classified as fragility fractures.
