Pos1432 the Impact of Prednisolone Starting Dose on Short and Long-Term Clinical Outcomes in Patients With Polymyalgia Rheumatica: The Retrospective Observational Cohort Study

Background: The prednisolone (PSL) dose ranging from 12.5 to 25 mg/day was recommended as an initial starting dose for the treatment of polymyalgia rheumatica (PMR) [1]. However, the evidence on the optimal initial PSL dose for PMR is scarce. Objectives: To assess the impact of PSL starting dose > 10 mg/day on early clinical remission, relapse after clinical remission, and PSL-free remission of PMR. Methods: The patients with PMR from two large tertiary medical care centers were included in this retrospective observational cohort study. The included patients fulfilled the 2012 EULAR/ACR provisional classification criteria for PMR and did not have any other rheumatic disease diagnosis throughout their clinical follow-ups. We defined "clinical remission" as clinical status with no musculoskeletal symptoms and CRP ≤ 0.3 mg/dL on PSL treatment. "Relapse" was defined as a clinical status requiring higher PSL or the addition of immunosuppressants (IS) to treat PMR-related symptoms after achieving clinical remission. "PSL-free remission" was defined as the clinical status with no relapse after stopping PSL, regardless of the concomitant use of the IS. The data on demographics and clinical findings were retrieved from electronic medical records. We performed logistic regression analysis for clinical remission rate within 4 weeks from the beginning of PSL treatment, relapse rate after achieving clinical remission, and PSL-free remission rate. We included age, sex, diabetes mellitus, C-reactive protein level at PMR diagnosis, duration (months) from the symptom onset to the beginning of the PSL treatment, the use of PSL starting dose > 10 mg/day, the concomitant use of IS, and the presence of relapse after clinical remission as explanatory variables. The last two variables were used only in the multivariate analysis for PSL-free remission. Results: Two hundred twenty-two patients (average age, 73.7 ± 9.0; female, 66.7%; median observation period, 26.3 [17.1, 43.3] months) were included. The numbers of patients on different PSL starting doses were 3 (1.35%) for 5 mg/day, 72 (32.4%) for 10 mg/day, 2 (0.90%) for 12.5 mg/day, 135 (60.1%) for 15 mg/day, and 10 (4.5%) for 20 mg/day, with 147 (66.2%) receiving PSL > 10 mg/day. Clinical remission rates at 4, 12, and 24 weeks were 49.1%, 86.0%, and 93.7%, respectively. Ninety-five (42.8%) relapsed, with 37 (16.7%) subsequently treated with IS (MTX,n=16; SASP,n=5; iguratimod,n=9; TCZ,n=2; sarilumab,n=1; TCZ+MTX,n=2; TCZ+iguratimod,n=1; sarilumab+iguratimod, n=1). PSL-free remission occurred in 112 (50.5%), with the median duration from the beginning of the treatment being 515 [336, 809] days. By logistic regression analysis, PSL starting dose > 10 mg/day showed an odds ratio (OR) of 2.16 (95%CI, 1.19-3.92: p=.011) for clinical remission within four weeks, an OR of 0.80 (95%CI,0.45-1.43: p=.45) for relapse after clinical remission, and an OR of 1.04 (95%CI,0.56-1.93: p=.91) for PSL-free remission. Logistic regression analysis on PSL-free remission rate showed an OR of 0.29 (95%CI, 0.14-0.56: p <.001) for the presence of relapse after clinical remission, an OR of 0.814 (95%CI, 0.712-0.931: p =.003) for the longer duration (months) from symptom onset to the beginning of PSL treatment, an OR of 0.946 (95%CI, 0.377-2.38: p=.906) for the addition of IS, and an OR of 1.65 (95%CI, 0.87-3.13: p=.125) for male sex, respectively. Conclusion: PSL starting dose >10 mg/day in PMR significantly correlated with early clinical remission rate but did not influence PMR relapse and PSL-free remission rates. REFERENCES: [1] Arthritis Rheumatol. 2015 Oct;67(10):2569-80. Acknowledgements: NIL. Disclosure of Interests: None declared.