092. RISK OF FRAGILITY FRACTURE OVER 10 YEARS IN POLYMYALGIA RHEUMATICA AND GIANT CELL ARTERITIS: A UK POPULATION STUDY

Abstract

Background: Polymyalgia rheumatica (PMR) and Giant Cell Arteritis (GCA) affect 1‐2% of the population aged over 55. There is no current epidemiological evidence about fracture risk in PMR or GCA. The aim of this cohort study was to estimate the incidence rates and relative risk of fragility fracture among patients with PMR and GCA, and the influence of glucocorticoid use on this risk. Methods: Incident cases of PMR or GCA were identified from 1990 to 2004 within the UK primary care dataset (Clinical Practice Research Datalink). Individuals over the age of 40 years, with incident diagnoses of PMR and GCA were identified from 1990 to 2004. For each case, four age‐, gender‐ and GP practice‐matched controls were randomly selected. Patients were followed up until the earliest date of: death, transfer out of practice, study end (31 August 2015) or the first fracture. Incidence rates of fragility fracture (defined as hip, vertebral, humerus or radial) per 10,000 person‐years were calculated for each disease group and hazard rates were compared to non‐PMR/GCA patients using Cox regression models, adjusting for cumulative steroid dose, PPI, bisphosphonate and methotrexate use. Results were further stratified by cumulative glucocorticoid use, using the dose reduction in UK guidelines to calculate typical cumulative dose for PMR (2817.5mg) and GCA (5705mg). Results: 2673 cases of GCA and 12,136 cases of PMR were identified and followed up for a median of 9.13 and 9.27 years, respectively. The incidence rate of fracture was 148.05 (95% CI 141.16, 155.28) per 10,000 person‐years for PMR and 147.15 (132.91, 162.91) for GCA. Both conditions were associated with an increased risk of fracture [PMR adjusted HR 1.44 (95% CI 1.33, 1.56); GCA adjusted HR 1.55 (95% CI 1.32, 1.82)] compared to age/gender matched controls. As cumulative glucocorticoid dose increased the risk of fracture in PMR increased, although this was not seen in GCA [PMR: 0‐2817.5mg HR 1.14 (95% CI 0.91, 1.42), >2817.5mg HR 1.74 (95% CI 1.56, 1.94); GCA: 0‐5705mg 1.94 (95% CI 1.45, 2.60), >5705mg HR 1.61 (95% CI 1.29, 2.01)]. The incidence rate ratio was raised 1 year after diagnosis [PMR: IRR 1.37 (95% CI 1.14, 1.66); GCA: IRR 1.76 (95% CI 1.13, 2.74)] and remained significantly raised 5 years after diagnosis [PMR: IRR 1.19 (95% CI 1.10, 1.28); GCA: IRR 1.18 (95% CI 1.01, 1.37)]. Conclusion: This study reports for the first time, an increase in fracture risk in patients with GCA and PMR which appears to be in part, independent of glucocorticoids cumulative dose and remains present 5 years after diagnosis.