Objectives: Antiepileptic drugs (AED) are known to increase the risk of low energy fractures by causing secondary osteoporosis. In this study, we determined the value of bone mineral density (BMD) and trabecular bone score (TBS) as predictors for low energy fractures in patients prescribed AED.

Material and Methods: This case-control study included 517 low energy fractures, i.e. a combination of ICD-10 codes according to the Swedish National Board of Health and Welfare's definition of osteoporosis- related fractures and ICD-10 codes for low energy trauma, in 1 385 subjects who had all participated in a TBS convertible DXA examination between July 2011 and November 2018 in Kalmar county, southeast Sweden. The AED-users were stratified according to Cytochrome P-450 enzyme inducing ability and compared to controls. The dataset was a combination of previously collected data from the electronic health register (EHR) and new BMD and TBS data from a DXA database. Univariate and multivariate logistic regressions were applied. 

Results: In this study BMD-spine did not predict low energy fractures; however, both BMD-hip (aOR: 0.35, 95% CI: 0.24-0.50) and TBS (aOR: 0.65, 95% CI: 0.46-0.92) were inversely related to fractures. Patients prescribed enzyme inducing AED had an increased fracture odds (aOR: 2.15, 95% CI: 1.24-3.72). There was no difference for fracture outcome between subjects with non-inducing drugs and controls (p-value 0.476). In this study 67.5% of the TBS convertible DXAs were performed within a year of the low energy fracture. Of the included patients, 82% were women.

Conclusions: While BMD-spine was not associated with low energy fractures, both BMD-hip and TBS predicted low energy fractures independently in patients prescribed AED. This is the first study demonstrating the value of TBS in patients prescribed AED. In clinical practice, BMD-hip and TBS could be combined to improve the assessment of bone health in this growing patient group. The value of the combination of the two measures should be assessed in a future prospective randomized study.