Trends in fractures in patients with spondyloarthritis: a nationwide population-based study (TREND-EspA study)

Our aim was to analyze trends in fracture rates in SpA patients over an extended time period. Only an increase of axial fractures, more specifically vertebral fractures, is observed in SpA. Purpose To analyze fracture incidence and trend in patients with spondyloarthritis (SpA) over an extended time period. Methods Retrospective observational population-based study with matched cohorts. Data from the Minimum Basic Data Set (MBDS) of Spain were reviewed. All SpA patient hospitalizations reported from 1999 to 2015 (SpA cohort) were analyzed. A control cohort (non-SpA cohort) matched by age, sex, region, and year of hospitalization was recruited. The age and sex-adjusted crude incidence rate was calculated for any fractures (axial and peripheral). Generalized linear models (GLM) were used for trend analysis. Association between fracture type and SpA (and its subtypes) was assessed using unconditional logistic regression models. Results In the SpA cohort, the age and sex-adjusted rates per 100,000 inhabitants/year of total fracture and different types of fracture were 45.72 any fractures, 17.64 axial, and 28.02 peripheral; 29.42 osteoporotic (12.67 vertebra, 12.29 hip, 1.50 pelvis, 1.82 humerus and 2.09 radius). In the non-SpA cohort, they were 65.79 any, 12.08 axial, 51.52 peripheral; 31.17 osteoporotic (4.94 vertebra, 16.15 hip, 2.29 pelvis, 3.64 humerus, 5.38 radius). Between 1999 and 2015, the trend in incidence rate for total fracture and different types of fracture increased similarly for both cohorts. In the SpA cohort, an increase of axial fractures was found (AOR 1.444; 95%CI 1.297-1.609), and specifically of vertebral fractures (AOR 2.440; 95%CI 2.097-2.839). Other types of fractures did not increase. Conclusions Only an increase of axial fractures, more specifically vertebral fractures, is observed in SpA. Trend in incidence is similar in both cohorts.