RESUMEN
OBJECTIVE: To evaluate the incidence and associated factors of initial and recurrent severe infections in hospitalized patients with systemic lupus erythematosus (SLE). METHODS: SLE patients that first hospitalized between 2010 and 2021 were studied retrospectively and divided into SLE with and without baseline severe infection groups. The primary outcome was the occurrence of severe infection during follow-up. Cox regression models were used to calculate the hazard ratio (HR) and 95% confidence interval (CI) for initial and recurrent severe infections. RESULTS: Among 1051 first hospitalized SLE patients, 164 (15.6%) had severe infection on admission. During a median follow-up of 4.1 years, 113 (10.8%) patients reached severe infection outcomes, including 27 with reinfection and 86 with initial severe infection (16.5% vs. 9.7%, p = .010). Patients with baseline severe infection had a higher cumulative incidence of reinfection (p = .007). After adjusting for confounding factors, renal involvement, elevated serum creatinine, hypoalbuminemia, cyclophosphamide, and mycophenolate mofetil treatment were associated with an increased risk of severe infection, especially initial severe infection. Low immunoglobulin, anti-dsDNA antibody positivity, and cyclophosphamide use significantly increased the risk of recurrent severe infection, with adjusted HR (95% CI) of 3.15 (1.22, 8.14), 3.60 (1.56, 8.28), and 2.14 (1.01, 5.76), respectively. Moreover, baseline severe infection and low immunoglobulin had a multiplicative interaction on reinfection, with adjusted RHR (95% CI) of 3.91 (1.27, 12.09). CONCLUSION: In this cohort of SLE, patients with severe infection had a higher risk of reinfection, and low immunoglobulin, anti-dsDNA antibody positivity, and cyclophosphamide use were independent risk factors for recurrent severe infection.