Background/Purpose: Patients with rheumatoid arthritis (RA) who have muscle weakness and stiff or painful joints might be at increased risk of falls and fractures. The present study aimed to prospectively determine the incidence of clinical fractures and associated predictors in patients with RA who participated in the TOMORROW study (UMIN000003876) that started in 2010.

Methods: We evaluated anthropometric parameters, bone mineral density (BMD), disease activity, RA medication, and the incidence of clinical fractures over a five-year period in 202 patients with RA (mean age, 58.6 y; medication with biological agents, 54.9%) and 202 age- and sex-matched healthy volunteers (controls; mean age, 57.4 y). We compared the incidence of clinical fractures between patients and controls between 2010 and 2015 and analyzed associated predictors in the patients using cox proportional hazard regression analysis.

Results: The incidence of clinical fractures did not significantly differ between patients with RA (0.042/person-years; py) and controls (0.034/py) within the five-year period (p = 0.35). And also, there were no difference in fractures sites between the two groups. Multivariable cox proportional hazard regression analysis adjusted for confounding factors including age, sex, smoking, and body mass index revealed that low BMD of the thoracic vertebrae (< 0.7 g/cm²) at entry was significantly associated with the incidence of clinical fractures (hazard ratio [HR], 2.63; 95% confidence interval [CI], 1.49 to 4.66; p = 0.001) in all participants group (Table 1). Although medication with glucocorticoid (GC) at entry was also a significant risk factor for fractures (HR, 2.14; 95% CI, 1.24 to 3.68; p = 0.006), RA morbidity was not (HR, 1.22; 95% CI, 0.74 to 2.01; p = 0.437). Among patients with RA, low BMD of the thoracic vertebrae (< 0.7 g/cm²) at entry was the most prominent risk factor for fractures (HR, 3.53; 95% CI, 1.52 to 8.15; p = 0.003). Additionally, medication with GC at entry (HR, 2.46; 95% CI, 1.28 to 4.73; p = 0.007) was a significant risk factor for fractures in the patients. A mean GC dosage of ≥ 2 mg/day during the five-year period increased risk for fractures in the patients (HR, 2.67; 95% CI, 1.06 to 6.72; p = 0.037).

Conclusion: There were no difference in the incidence of clinical fractures between patients with RA and controls during a period of five years. Low BMD of the thoracic vertebrae and low GC doses (≥ 2 mg/day) are apparently significantly associated with the incidence of clinical fractures among patients with RA.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/glucocorticoid-use-and-low-thoracic-bone-mineral-density-are-predictors-for-clinical-fractures-in-patients-with-rheumatoid-arthritis-five-year-findings-of-the-tomorrow-study/