Regional fat distribution as a novel predictor of fracture risk in inflammatory bowel disease: a DXA-based cohort study
Suvan Suntharalingam, Marwan BukhariObjective
To determine whether dual-energy X-ray absorptiometry (DXA)-derived femoral and abdominal fat percentage was independently associated with fracture history in adults with inflammatory bowel disease (IBD), with a focus on the glucocorticoid-treated IBD population.
Methods
A retrospective cohort study of adults with a confirmed diagnosis of IBD underwent clinically recommended DXA scanning to quantify abdominal and femoral fat mass, bone mineral density (BMD) and total body composition between June 2004 and February 2024 in northwest England. Fracture history was the main outcome. After controlling for hip lean mass, body mass index (BMI), BMD, comorbidities and glucocorticoid exposure, multivariable logistic regression models assessed relationships between regional fat percentages and fracture history. To determine if the association was independent of peripheral lean mass, a subgroup analysis of 300 glucocorticoid-exposed patients and a sensitivity model incorporating hip lean mass were performed.
Results
A total of 1302 patients with IBD underwent DXA scanning. Abdominal fat percentage showed a positive correlation throughout the full cohort (OR 1.02, 95% CI 1.001 to 1.037). The correlation remained in patients receiving glucocorticoids (OR 1.06, 95% CI 1.01 to 1.11) and in the lean-mass sensitivity model (OR 1.05, 95% CI 1.01 to 1.09). In all models, there was no correlation between the femoral fat percentage and fracture history. In the full cohort, low BMI was linked to fracture history (OR 2.48, 95% CI 1.02 to 6.04), although BMI (continuous) was only negatively correlated with glucocorticoid exposure (OR 0.91, 95% CI 0.84 to 0.99). After controlling for hip lean mass, BMI and BMD, and glucocorticoid exposure, the link between abdominal fat remained strong.
Conclusion
The only regional body composition parameter that was consistently linked to fracture history in patients with IBD, including those receiving glucocorticoids, was abdominal fat percentage. This association held regardless of BMI, BMD and peripheral lean mass. According to these findings, skeletal vulnerability in IBD is linked to this phenotype of abdominal adiposity. To assess potential clinical value and show temporal correlations, prospective studies are required.