Endocrine Abstracts (2017) 51 OC4.2 | DOI: 10.1530/endoabs.51.OC4.2

Cumulative radiation exposure from imaging and associated lifetime cancer risk in children with osteogenesis imperfecta

Amy Thorby-Lister1, Wolfgang Högler1,2, Kirsten Hodgson3, Nicola Crabtree1, Nick Shaw1 & Vrinda Saraff1


1Department of Endocrinology and Diabetes, Birmingham Children’s Hospital, Birmingham, UK; 2Institute of Metabolism and Systems Research, University of Birmingham, Birmingham, UK; 3Radiation Protection Services, University Hospitals Birmingham, Birmingham, UK.


Background and objectives: Children with Osteogenesis Imperfecta (OI) require frequent imaging for fractures and surveillance of bone mineral density. Radiation exposure in childhood carries the highest risk of cancer. Here, we estimate the cumulative effective radiation dose (E) and lifetime cancer risk (LAR) from imaging in children with OI. We also explore the hypothesis that the rate of fracture positive imaging for investigation of injuries is related to family history of OI.

Methods: We reviewed all imaging (X-ray, computed tomography [CT] & bone densitometry [DXA]) conducted from 2003 to 2016 in children with OI (0–19 years) with a minimum observation period of 5 years, at a tertiary paediatric centre. E was estimated for each image using age-dependent local reference data. LAR was calculated using cumulative E with organ, sex and age specific risk coefficients. Patients were then categorized by cancer risk.

Results: We present results from 106 children (50% females, 5747 images) with mild (n=74), moderate (n=22) and severe (n=10) OI. The median (range) observation period was 11.7 years (5.2–15.6). The number of images taken per year for mild OI was 3.1 (1.9–5), moderate 5.4 (2.7–8.2) and severe 9.8 (5.9–22.9). CT accounted for 0.8% of total imaging episodes but contributed 66% of total E. Cumulative E and LAR differed significantly between mild and moderate OI (P=0.006), and mild and severe (P=0.001). The median LAR of cancer was 8.8 cases per 100,000 exposed patients (0.8–403), which increased with OI severity. Of the patients with a moderate risk of cancer all had undergone CT scans and 88% had scoliosis or vertebral fractures. Family history of OI did not impact cumulative E, LAR or rate of fracture positive imaging, which was 60%.

Conclusions: When compared to baseline LAR (50%) the additional cancer risk in our OI cohort, from imaging, was small (0.0088%). Those with vertebral fractures, scoliosis or severe OI carried the highest risk of cancer, and in this group the LAR reached 0.4%. We recommend replacing spinal x-rays with vertebral fracture assessments on DXA (7.5 times less radiation), and exercising caution with CT imaging to further minimise cancer risk.

Article tools

My recent searches

No recent searches.