To estimate the cumulative effective dose of radiation (E) and additional lifetime attributable risk (LAR) of cancer from ionizing radiation in children with osteogenesis imperfecta (OI), who require frequent imaging for fractures and bone densitometry (DXA) surveillance. Also, to evaluate the pattern of long bone fractures.
We reviewed all imaging (x-rays, DXA and computed tomography [CT]) conducted in a cohort of children with OI with a minimum observation period of 5 years. For each image, E was estimated using age-dependent local data, and LAR of cancer was extrapolated. LAR and fracture data were compared among children with mild, moderate and severe OI. LAR was allocated to cancer risk categories, and the moderate risk group (1 in 1000 to 1 in 100) was evaluated further.
Results from 106 children with OI (50% females, 5747 images) are presented, with a median (range) observation period of 11.7 (5.2-15.6) years. CT accounted for 0.8% of total imaging procedures but contributed to 66% of total E. The overall LAR of cancer was minimal, averaging an additional 8.8 cases per 100,000 exposed patients (0.8-403). LAR was significantly lower in children with mild OI compared to those with moderate (p = 0.006) and severe OI (p = 0.001). All patients with a moderate LAR of cancer (n = 8) had undergone CT scans and 88% had scoliosis or vertebral fractures. The cohort experienced 412 long bone fractures, with the most common site being the femur (26.5%). OI severity correlated positively with long bone fracture rates (p < 0.001).
When compared to baseline LAR of cancer (50%) the additional cancer risk from ionizing radiation imaging in our paediatric OI cohort was small (0.0088%). To reduce additional cancer risk, we recommend replacing spinal x-rays with vertebral fracture assessments on DXA and exercising caution with CT imaging.