Occurrence, predictors and outcome of infections at three months in hospitalized patients with SLE: A prospective study from Southern India

Abstract

Prospective data on infections in systemic lupus erythematosus (SLE) from India are scarce. We studied the frequency and predictors of infections in hospitalized SLE patients. All data on infections were prospectively recorded. During the study period, 212 SLE patients (91% women) were hospitalised. Sixty-three (29.7%) had infections. The most common infections were pneumonia, skin and soft-tissue infections and urinary-tract infections. Mortality was higher in the infection group compared to the no infection group (11.1% vs. 0.7%; p=0.01). At three months, 10/63 developed another episode of infection. On logistic regression, the predictors of infection were fever [odds ratio (OR) = 4.17], vasculitis (OR = 2.64), thrombocytopaenia (OR = 3.59), presence of co-morbidities (OR = 3.59) and duration of hospital stay >11 days (OR = 3.55); p<0.001 for all. High-sensitivity CRP (hsCRP) and procalcitonin were measured in 95 patients. hsCRP was significantly higher in the infection group compared to the no infection group (median 27 vs. 6.5 mg/L; p<0.001). Procalcitonin was similar in both groups. A hsCRP of 11.5 mg/L had a sensitivity of 66.7% and specificity of 72.9% to diagnose infection. To conclude, 29.7% of Indian SLE patients who are hospitalized have infections, resulting in a significantly increased duration of hospital stay. The presence of fever and co-morbidities are predictors of infection. The role of high hsCRP to support the diagnosis of infection in SLE is once again confirmed.