P416 Use of DXA in children with Inflammatory Bowel Disease – a large single centre study

Jois, A.(1);Perera, S.(2);Simm, P.(3);Alex, G.(1);

(1)The Royal Children's Hospital, Gastroenterology and Clinical Nutrition, Melbourne, Australia;(2)The Royal Children's Hospital, General Medicine, Melbourne, Australia;(3)The Royal Children's Hospital, Endocrinology and Diabetes, Melbourne, Australia


Low bone mineral density (BMD) is a complication in children with Inflammatory Bowel Disease (IBD). Dual-energy X-ray absorptiometry (DXA) is an established screening tool for BMD, yet there are limited data to guide its use in children with IBD. We performed a single site retrospective analysis of the use of DXA and evaluated factors associated with low BMD.


Children 3-18 years with IBD diagnosed between 2013-2018 at The Royal Children’s Hospital, Melbourne, Australia, were included. Patient, disease and treatment demographics were collected alongside vitamin D, calcium, CRP, ESR, albumin and faecal calprotectin (FCP) (averaged over 6 months before and after DXA). Rates of corticosteroid use, vitamin D and calcium supplementation, bisphosphonate use and fractures were also collected. Mann-Whitney and Fisher exact tests were used for continuous and categorical group comparison, respectively. Data are presented as median (interquartile range).


239 children diagnosed at a median age of 12 (9.1-14.2) were followed for 5.1 (4-6.4) years. 72/239 (30%) children had at least one DXA at 11 (1.25-28.8) months post diagnosis. 28/72 (39%) children had a follow-up DXA 2.3 (1.9-2.9) years post diagnosis. Children referred for DXA had a lower weight centile (48.7 (17.1-78.2) vs 59.7 (31.4-84.7), p=0.03), and were more likely to have Crohn’s disease (OR 2.18, p=0.01). At first DXA, median lumbar spine (LS) Z score was -0.80 (-1.65-0.08), height adjusted LS Z score was -0.65 (-1.18-0.10), hip Z score was -1.30 (-1.80--0.35) and total body less head Z score was -1.40 (-2.55--0.70). 18/72 children had LS Z score > 0. Children with LS Z score < -2.0 (n=14) had lower weight (6.57 (1.78-23.7) vs 51.1 (26.5-68.7), p=0.0002) and height centiles (3.62 (1.17-17.1) vs 42 (16.9-67.1), p<0.0001), higher FCP (3041 (1182-4192) vs 585 (139-2419), p=0.009) (Figure 1), and higher odds of calcium supplementation (OR 16 (95% CI 2.93-89.1), p=0.003) and endocrinology review (OR 9.61 (95% CI 2.65-31.0), p=0.001). No fractures were reported. Of the 28 children with serial DXAs, there was no significant change in Z scores. When comparing children with a worse (16/28) Z score at second DXA to those with improvement (12/28), there was a trend toward lower vitamin D levels (37 (25.2-58.3) vs 62 (46.3-87), p=0.06) at first DEXA, and higher ESR at second DEXA (23.4 (19.3-29) vs 8.20 (3.5-21.9), p=0.02).


Almost one third of children with IBD at a tertiary referral centre underwent DXA, with lower BMD than age and sex matched controls. Disease activity markers FCP and ESR were associated with lower BMD. There was no significant change in DXA score over time. No fractures were identified over the study period.