Bacteremia Associated With Severe Flares in Systemic Lupus Erythematosus

Investigators abstracted data from RELESSER, a registry of 3679 patients with SLE from 45 Spanish hospitals. All bacteremic episodes identified via RELESSER were included in subsequent analyses. Clinical and laboratory characteristics concerning bacteremia and SLE status were extracted, as was information about patient comorbidities at the time of bacteremic infection. Patients were age- and sex-matched with SLE controls without bacteremia. Logistic regression analyses were conducted to identify differences in SLE characteristics across patients.

A total of 114 episodes of bacteremia in 83 patients were identified, for an overall incidence rate of 2.7 cases per 1000 patient-years. The median age at time of bacteremia infection was 40.5 years, and 88.6% of patients with bacteremia were women. Among patients with bacteremia, median SLE duration was 9.7 years and the median Systemic Lupus Erythematosus Disease Activity Index score was 4. Additionally, 41.2% of those with bacteremia had a coincident SLE flare and 66% of these were severe flares. SLE was serologically active in 50.9% of cases, and some comorbidity was observed in 64% of cases.

Regarding SLE treatments at the time of bacteremia, 88.6% of patients received corticosteroids (68.6% >10 mg/d) and 57% received immunosuppressors. Bacteremia was caused by gram-negative bacilli in 52.6% of cases, most frequently Escherichia coli (29.8%). Bacteremia-related mortality was 14%, and infection was recurrent in 27.2% of cases.

In multivariate analyses adjusted for disease duration, only elevated creatinine (odds ratio [OR], 1.31; 95% CI, 1.01-1.70; P =.045), diabetes (OR, 6.01; 95% CI, 2.26-15.95; P =.000), cancer (OR, 5.32; 95% CI, 2.23-12.70; P =.000), immunosuppressors (OR, 6.35; 95% CI, 3.42-11.77; P =.000), cyclophosphamide use (OR, 9.37; 95% CI, 5.12-17.14; P =.000), and damage (OR, 1.65; 95% CI, 1.31-2.09; P =.000) remained statistically significant in their association with bacteremia. A dose-response relationship was observed between corticosteroids and bacteremia risk.

Increased flare severity and corticosteroid use, along with other comorbidities, were identified as predictors of bacteremia infection in patients with SLE. Clinicians may find this information useful in projecting infection risk among vulnerable patients.

Reference

Rúa-Figueroa I, López-Longo F, Del Campo V, et al. Bacteremia in systemic lupus erythematosus patients from RELESSER: risk factors, clinical and microbiological characteristics and outcomes [published online April 15, 2019]. J Rheumatol. doi:10.3899/jrheum.180882