Diagnosis of Recurrent Fracture in a Pediatric Cohort

Significant fracture history in children is defined as having at least one vertebral fracture, at least 2 fractures by age 10, or at least 3 fractures by age 19. Between September 2011 and December 2014, clinical data were collected on children with a significant fracture history that attended a major Australian children's hospital. Fifty-six patients were identified as having 305 fractures in total, including 44 vertebral fractures. 18% of patients (10/56) were diagnosed with osteogenesis imperfecta (OI) by a bone health expert, molecular testing or both, and they sustained 23% of all fractures (71/305). Analysis of serum bone biochemistry showed all median values to be within a normal range and no clinically significant differences between patients with and without OI. The DXA and pQCT derived bone mineral density (BMD) and bone mineral content (BMC) Z scores were reduced overall. DXA derived total body and lumbar spine areal BMD-for-age and BMC-for-age Z scores were significantly lower in children who had vertebral fractures or who were later diagnosed with OI. Similarly, pQCT performed on radii and tibiae showed Z scores significantly less than zero. pQCT-derived limb muscle cross sectional area Z scores were significantly lower in the OI subgroup. In conclusion, this study describes the bone phenotype of children referred to a tertiary hospital clinic for recurrent fractures and highlights a subset of children with previously undiagnosed OI, but a larger cohort without classic OI. Thus it can be clinically challenging to differentiate between children with OI type 1 (mild phenotype) and non-OI children without bone densitometry and genetic testing. We conclude that recurrent fractures in children should prompt a comprehensive bone and systemic health assessment to eliminate an underlying pathology.