Background Fragility fractures are a consequence of poor bone health, this is especially seen in patients on corticosteroids. Polymyalgia rheumatica (PMR) is usually treated with corticosteroids for a median of two years (1). The predictors of fragility fractures in this cohort have not been explored in a large observational cohort. Furthermore it is unknown which other risk factors for Osteoporosis contribute to the risk of fractures in this cohort The FRAX™ tool uses the BMD in the femoral neck as a predictor for fracture, the ability of the lumbar spine bone mineral density (BMD) to increase prediction of fragility fracture has not been explored.
Objectives To explore the further predictors of fragility fractures in patients with PMR on corticosteroids and to explore the added predictive value of lumbar spine BMD.
Methods Patients referred for BMD estimation using a dual X-ray absorptiometry (DEXA) machine to a scanner in the North West of England between 2004 and 2014 were included in the study. Patients with a history of PMR on steroids were extracted. Initially those that sustained a fracture were compared to those that had not sustained a fracture using the chi squared test for categorical variables and Students T test for continuous variables. The predictors of fracture in that cohort were explored using a logistic model, initially in a univariate manner, and subsequently in a stepwise logistic model. Predictors included age at scan in years, gender, smoking, excess alcohol consumption, the presence of rheumatoid arthritis, body mass index (BMI) and family history of fragility fracture in addition to BMD in the lumbar spine and femoral neck.
Results 703 patients were included in the analysis, 525 (74.7%) were female. Mean age was 70.4 years (SD 8.9 years). 154 (21.9%) had sustained a fracture. Comparing the patients who had and who had not sustained a fracture showed that they were more likely to be female 138/525 (26%) vs 16/178 (9%) (p<0.001), older 72.5 years (SD9.0) vs 70.0 (SD8.8) p=0.002 and have a lower BMD in the lumbar spine and the femoral neck. The results of the logistic models unadjusted and adjusted for age are shown in table 1 below with significant values denoted with an asterisk (*)
In the multivariate model aside from age, female gender (OR 2.010 95%CI 1.11,3.99) and femoral neck but not lumbar spine BMD remained in the model (OR 0.07 95%CI 0.01,0.59), surprisingly family history of fracture (OR 2.21 95%CI 1.01,4.84) and a higher BMI (OR 1.06 95%CI 1.01,1.10) also were significant predictors.
Conclusions In this cohort of patients the females, older patients and low BMD in the femoral neck were all associated with fractures, which is not unexpected, however added risks were identified which are not necessarily clinically obvious. BMD in the lumbar spine did not add to the predictive value of sustaining a fracture. Further prospective work is needed to validate these findings.
Disclosure of Interest None declared