Objective:To determine the characteristics of hospitalized newly diagnosed systemic lupus erythematosus (SLE) patients with high disease activity, and identify the risk factors.Methods:Data from 194 newly diagnosed SLE patients at Shanghai Renji Hospital between May 2013 and December 2018 were collected retrospectively. All patients were followed up for 1 year or until death. Patients' demographic, clinical, and laboratory characteristics on admission and medication history were retrospectively collected as baseline data. Patients were divided into two groups, lupus patients with infection (51 cases) and lupus patients without infection (143 cases). The method of univariate analysis of data depended on the data distribution type. Variables that suggested association in the univariate analysis ( P<0.05) were entered into Cox regression model. Results:Among 194 patients with newly diagnosed SLE, 21 cases (11%) died and 51 cases (26%) were infected during 1-year follow-up. Regarding the infection site, 34 cases (67%) had lung infection, 9 cases (18%) had central nervous system infection and 9 cases (18%) had blood stream infection. Common bacteria were identified in 19 cases (45%), followed by fungal infection in 18 cases (43%) and mycobacterium infection in 7 cases (17%). Among the 51 patients with infection, 38 patients (75%) had infection within the first 3 months after diagnosis, and mortality in this group was significantly higher than that in the uninfected group (39%, 15/38 vs 2%, 3/143 , P<0.01). Comparing baseline parameters between patients with 3-month infection and without, significant differences ( P<0.05) were detected in age (≥40 years), systemic lupus erythematosus disease activity index (SLEDAI) score (>10 points), Systemic Lupus International Collaborating Clinic (SLICC)/American College of Rheumatology(ACR) systemic lupus erythematosus damage index (SDI) (≥1 point), pericardial effusion, nephritis, gastrointestinal vasculitis, diabetes, lymphocyte count <0.8×10 9/L platelet count <100×10 9/L, serum creatinine >104 mmol/L and serum globulin level <20 g/L. Finally, clinically meaningful candidate predictors were included in the Cox regression model and it showed that lymphocyte count <0.8×10 9/L, nephritis and gastrointestinal vasculitis were independently predictive for 3-month infection in new-onset lupus patients. Conclusion:Understanding disease spectrums and risk factors of infection in newly diagnosed SLE patients will help clinicians to manage those patients with infection effectively to achieve favorable prognosis.

Objective:To evaluate the association between the efficacy and safety of metformin and the influence of variants in SLC47A1 rs2289669 G>A polymorphism in the treatment of systemic lupus erythematosus (SLE).Methods:A multicenter, randomized, double-blind, placebo-controlled trial was conducted. Patients were consented at enrollment for blood donation for genotyping, and their peripheral blood were used to detect the distribution frequency of SLC47A1 mutations. The major or mild/moderate flares defined by modified safety lupus erythematosus national assessment (SELENA)-systemic lupus erythematosus disease activity index (SLEDAI) Flare Index (SFI) and adverse events were recorded at 12 months of follow-up. The correlation between efficacy/safety and genotype was analyzed. Student's t test and χ2 test was used to assess the continuous variables and categorical variables. Results:Between May 24, 2016, and Dec 13, 2017, a total of 31 patients in the metformin group and 35 in the placebo group were detected. There were no statistical significant differences in the clinical manifestations, SELENA-SLEDAI scores, and therapy of the participants at baseline. There was no significant difference in the frequency of AA genotype, GA genotype, and GG genotype of SLC47A1 rs2289669 distribution between the metformin group and the placebo group. In the metformin group, patients who flared had a lower frequency of A alleles than those non-flared [25%(4/16) vs 61%(28/46), χ2=6.116, P=0.019 8]; the flare rate was significantly lower in patients with AA genotype than in GG genotype [0%(0/8) vs 57%(4/7), χ2=6.234, P=0.012 5]. The infection rate was lower in the metformin group than that in the placebo group [38%(12/31) vs 69%(24/35), χ2=5.913, P=0.015 0], but there was no significant difference among different genotypes in the metformin group. Compared to GG geno-type, AA genotype showed a trend of decrease in infection rate[38%(3/8) vs 72%(5/7), χ2=1.727, P=0.188 8]. Conclusion:Metformin has a favorable safety profile and may reduce the frequency of flares in SLE patients with low-grade lupus disease activity. The metformin therapeutic efficacy in SLE is relevant to the SLC47A1 gene polymorphism. Patients of the AA genotype may benefit most from metformin than those of the GG and GA genotypes.