USEFULNESS OF THE TRABECULAR BONE SCORE AS A PREDICTOR OF VERTEBRAL FRACTURE IN PATIENTS WITH AXIAL SPONDYLOARTHROPATHY

Background: In axial spondyloarthritis (axSpA) the risk of vertebral fracture is increased, not always corresponding with the values of bone mineral density (BMD). One possible explanation is that syndesmophytes interfere with these values. We consider whether the evaluation of trabecular microarchitecture by an accessible methodlike the Trabecular Bone Score (TBS), that does not involve additional irradiation neither seem to be influenced by the presence of syndesmophytes, may be an advantage to estimate the risk of fracture. Objectives: To estimate the prevalence of vertebral fractures in patients with axSpA. To assess the diagnostic accuracy of TBS and BMD for vertebral fracture, and if it is influenced by the presence of syndesmophytes. To analyze the correlation between the absolute values of BMD and TBS in the lumbar spine. Methods: Cross-sectional study. Patients were consecutive recruited. We collected demographic (sex, age), clinical (syndesmophytes, vertebral fracture, BASDAI, BASFI, time of evolution of axSpA, treatment) and analytical variables [vitamin D (1,25-OHD), CRP and ESR]. The BMD was determined using the Lunar Prodigy ProTM densitometer from GE Healthcare, to which the TBS iNsight® software version 2.2 was added to perform the TBS analysis. The presence of fracture was evaluated by radiology. The statistical analysis was performed with the SPSS 22.0 and OpenEpi softwares. Results: 84 patients were included, 60 men and 24 women, with a mean age of 59 years (± SD 13). 51.2% had lumbar syndesmophytes. The prevalence of fractures was 13.7%, 95 CI (7.8-22.9). 51.2% were treated with NSAIDs, and 48.8% with biological drugs. The evolution of axSpA was \u003e 10 years in 65.5%. The mean scores of BASDAI and BASFI were 3.7 and 4.3 respectively (± SD 2.2 and 2.3). The mean CRP value was 8.5 mg / L (± SD 8.4), ESR 12.2 mm / h (± SD 11.4) and 1.25-OHD 27.9 ng / dL (± SD 13.6). According to the lumbar and femoral T Score, 9.5% and 15.5% of the patients were in the range of osteoporosis respectively.19% patients had a low TBS value (≤1.23). Regarding the influence of syndesmophytes on TBS and BMD values, we found significant differences in lumbar spine BMD (p = 0.01) but not in total hip and femoral neck BMD (p = 0.2 and 0.3 respectively) nor in the TBS (p = 0.1). Regarding the correlation of TBS and BMD values of the spine, no correlation was observed in patients with syndesmophytes, while a moderate correlation (r = 0.4, p = 0.02) was observed in patients without syndesmophytes. In the univariate analysis, the factors related to the presence of vertebral fracture were age, female sex, absolute BMD values in the lumbar spine and total hip, and TBS values. No relationship was found with the rest of the variables. In the multivariate analysis, only the TBS showed a significant association with the presence of fractures (p =0.02). Regarding the predictive capacity of fractures, TBS showed a higher sensitivity than that of BMD (55.6% versus 18.2% and 30% of BMD in the spine and hip respectively), being the specificity comparable (85.3% versus 91.3% and 85.1% of BMD in column and hip respectively). Conclusion: the prevalence of fractures was 13.7% among the patients studied, 95 CI (7.8-22.9). The presence of syndesmophytes influenced the values of lumbar BMD but not the hip BMD or those of the TBS. We found a correlation between the values of BMD of the spine and TBS only in patients who did not have syndesmophytes. Only TBS values were significantly related to the presence of fractures in the multivariate analysis. TBS showed greater sensitivity with similar specificity than BMD for the detection of vertebral fractures. Disclosure of Interests: Ana V Orenes Vera: None declared, L Montolio-Chiva: None declared, I Vazquez-Gomez: None declared, Eduardo Flores: None declared, Elia Valls-Pascual Grant/research support from: Roche, Novartis, and AbbVie, Speakers bureau: AbbVie, Lilly, Pfizer, MSD, Novartis, Janssen, Bristol Myers Squibb, UCB Pharma, A Martinez-Ferrer: None declared, Desamparados Ybanez: None declared, Luis Garcia-Ferrer: None declared, Maria Vega-Martinez: None declared, Magdalena Graells-Ferrer: None declared, A Sendra-Garcia: None declared, V Nunez-Monje: None declared, Inmaculada Torner Hernandez: None declared, Juanjo J Alegre-Sancho Consultant of: UCB, Roche, Sanofi, Boehringer, Celltrion, Paid instructor for: GSK, Speakers bureau: MSD, GSK, Lilly, Sanofi, Roche, UCB, Actelion, Pfizer, Abbvie, Novartis