POS0713 PREDICTORS OF HOSPITALIZATION IN PATIENTS WITH SYSTEMIC LUPUS ERYTHEMATOSUS: A 10-YEAR COHORT STUDY OF 398 PATIENTS FROM A TERTIARY CENTRE

Background:Patients with systemic lupus erythematosus (SLE) often require hospitalization. The cause of admission may vary, but active disease and infection are consistently reported as the main reasons for hospitalization and are associated with worse survival and damage accrual. Recent improvements in the standard of care, including minimization of glucocorticoid dose and more effective and safe immunosuppressive regimens, may have changed the incidence and risk factors for hospitalization due to these causes. Hence, it is useful to identify predictors of hospitalization to further reduce the risk of admission for disease activity and severe infection in patients with SLE.Objectives:To identify predictors of hospitalization in patients with SLE, according to the underlying cause.Methods:Patients with SLE fulfilling classification criteria (ACR’97 and/or SLICC), regularly followed at an academic lupus clinic from January 2009 to December 2020 and with at least two outpatient visits were included in this cohort study. Time to first hospitalization up to 120 months was identified separately for the following admission causes: (a) any cause; (b) active SLE; (c) infection. Predictors of hospitalization were sought through survival analysis, with distinct models for each of the major admission causes. Univariate analysis was performed using Kaplan-Meier curves and Log-Rank tests. Tested variables assessed at baseline included: gender; age at SLE onset; age; disease duration; SLE Disease Activity Index (SLEDAI-2K) score; ongoing antimalarial use; ongoing immunosuppressants; ongoing prednisolone daily dose; lupus nephritis up to baseline; SLICC Damage Index (SDI) score. Variables with p<0.1 were further tested in multivariate Cox regression models. Hazard ratios (HR) were determined with 95% confidence intervals (95%CI).Results:We included 398 patients (female: 86.2%, mean age: 41.2±15.1 years, mean disease duration: 10.1±9.2 years; previous lupus nephritis: 28.9%; mean SLEDAI-2K score: 3.4±2.7; ongoing antimalarials: 78.9%; ongoing immunosuppressant: 29.9%; ongoing prednisolone >7.5 mg/day: 17.1%; SDI score ≥1: 28.4%). During the follow-up period, 50.5%, 23.6% and 17.3% were hospitalized at least once for any cause, active SLE or infection, respectively.In the multivariate model, significant baseline predictors for hospitalization due to active disease were (table 1): SLEDAI-2K score >5; disease duration ≤2 years; ongoing immunosuppressants; SDI score ≥1. Baseline independent predictors of hospitalization for infection included (table 1): male gender; SDI score ≥1; ongoing antimalarials were protective.Table 1.Predictors of hospitalization in multivariate Cox regression according to the admission causePredictorsHospitalization for active SLEHospitalization for infectionSLEDAI-2K score >52.43 (1.53-3.88)n.s.SLE duration ≤2 years1.70 (1.04-2.77)n.s.Ongoing immunosuppressant1.91 (1.24-2.95)n.s.SDI score ≥11.82 (1.16-2.86)2.14 (1.33-3.45)Male gendern.s.2.19 (1.23-3.89)No antimalarial treatmentn.s.2.20 (1.34-3.60)Risk for each predictor reported as Hazard Ratio (95% Confidence Interval); n.s.: non-significantConclusion:Tight control of disease activity, prevention of damage accrual, and treatment with antimalarials may contribute to minimize the risk of hospitalization for these two major causes of admission in patients with SLE.Disclosure of Interests:None declared