Pos0441 Predictors of Giant Cell Arteritis in Patients With Polymyalgia Rheumatica in Southern Sweden: A Retrospective Study

Background: Polymyalgia Rheumatica (PMR) is an inflammatory condition affecting individuals aged 50 years and over, predominantly women. PMR sometimes coexists with Giant Cell Arteritis (GCA), the most commonly occurring vasculitis in adults in Europe and North America. Results of previous studies of occurrence GCA in patients with PMR have been variable, likely reflecting differences in study design. Objectives: To identify risk factors for GCA in patients with PMR. Methods: Patients from two population-based prospective health surveys of residents in Malmö in Sweden (the Malmö Diet and Cancer Study and the Malmö Preventive Medicine Project), with a subsequent diagnosis of PMR or GCA with PMR in the Skåne Healthcare Register or the Swedish National Patient Register were identified. Medical records were obtained from the regional hospital based and Primary Care based systems. All the available medical records were systematically reviewed for relevant data, and PMR diagnoses were verified by a specialist in rheumatology in an independent review. Descriptive data on baseline characteristics at the time of PMR diagnosis were obtained. Logistic regression was performed to determine possible risk factors for GCA at PMR diagnosis or later. Results: Records for a total of 1508 patients with a registered diagnosis of PMR were reviewed. Among 1030 patients with sufficient available data, the PMR diagnosis was verified in 61% (n=626). The mean age at PMR diagnosis was 75.8 (standard deviation 6.8) years; 63 % were women. GCA was diagnosed in 37 patients within one month after PMR diagnosis, and in 20 patients at a later time point. Female patients with PMR were more likely to have or develop GCA (odds ratio (OR) 2.38; 95% confidence interval (CI) 1.23-4.61). Higher ESR, CRP in the highest quartile and higher Prednisolone dose at PMR diagnosis were also associated with GCA, and there was a negative association with hip pain/stiffness (Table 1). In multivariable analysis, there were significant associations for female sex and higher Prednisolone dose with GCA (Table 1), and patients with stiffness or pain in the hip at PMR diagnosis were less likely to also be diagnosed with GCA (adjusted OR 0.47; 95 % CI 0.23-0.97). In analyses of predictors of GCA >1 month after PMR diagnosis, there was also a trend towards an increased risk in female patients, but no associations with laboratory markers of inflammation or baseline Prednisolone dose (Table 2). Patients with documented morning stiffness at PMR diagnosis had a lower risk of developing GCA after 1 month or more (OR 0.53; 95% CI 0.29-0.98). Conclusion: In a structured review of medical records from 626 patients with verified PMR, diagnosed in primary or specialized care, we found that 9% developed GCA at or after PMR diagnosis. Female patients were overrepresented among those diagnosed with both PMR and GCA. The lower proportion with GCA among those with hip symptoms, and the lower risk of later GCA among those with morning stiffness suggest that extensive musculoskeletal involvement and symptoms and signs of vasculitis constitute different parts of the PMR-GCA spectrum. When compared to patients with PMR only, starting dose of Prednisolone was higher in patients with coexisting GCA at or any time after PMR diagnosis although not in the subgroup that developed GCA more than one month after PMR. This likely reflects treatment decisions based on an initial clinical suspicion for GCA. REFERENCES: NIL. Acknowledgements: NIL. Disclosure of Interests: Charlotta Fors: None declared, Ulf Bergström: None declared, Aladdin J Mohammad: None declared, Carl Turesson AbbVie, Boehringer-Ingelheim, Nordic Drugs, Novartis, Pfizer, Roche.