Distinction between infection and flare in patients with systemic lupus erythematosus (SLE) is a challenge in clinical practice. Objective To analyze the utility of neutrophil-to-lymphocyte ratio (NLR) plus C-reactive protein (CRP) to differentiate between infection and active disease in patients with SLE. Methods A cross-sectional study of a cohort of patients with SLE was carried out. Blood samples from four groups (patients without infection or active disease, patients with infection, patients with active disease, and patients with both infection and active disease) before therapeutic interventions were analyzed. We excluded patients with current malignancy, pregnancy, ischemic heart disease or use of antimicrobials during previous 7 days. Hematological cell count, CRP and cultures were obtained. We constructed receiver operating characteristic curves; sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated. Results Forty patients were included. NLR cut-off ≥6.3 had sensitivity 70%, specificity 85%, PPV 83% and NPV 74% to detect patients with non-viral infections. A CRP cut-off ≥7.5 mg/L had sensitivity 90%, specificity 75%, PPV 78% and NPV 88% to detect infections regardless of SLE activity. Combination of CRP plus NLR improves the specificity to 90% and PPV to 88%. Excluding the group with both infection and active disease, CRP plus NLR expands specificity to 95% and NPV to 90%. Conclusion In our experience, levels of CRP, particularly CRP plus NLR, were useful in differentiating patients with SLE from those with suspected non-viral infection regardless of the activity of the disease.

No abstract