Diagnostic accuracy study of globular sedimentation rate and C-reactive protein in patients with systemic lupus erythematosus and fever admitted to a Social Security hospital in Lima, Peru, 2010-2019

Introduction: In patients with systemic lupus erythematosus (SLE) and fever, the causes are infection and/or activity. We assessed the diagnostic accuracy of C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) to discern these causes. Materials and methods: We reviewed medical records of patients with SLE and fever hospitalized from 2010-2019. Based on two categories of cause of fever (infection with-or-without activity versus activity alone), we computed ROC curves, areas under the curve (AUC) and best cut-off values for ESR, parameter ESR, CRP and ESR/CRP ratio. In addition, we calculated sensitivity, specificity, and predictive values. Results: We included 40 cases (34 women) with an average age of 35.5 years. The main cause of fever was the coexistence of infection and activity (17/40; 43%), followed by fever only due to activity (14/40; 35%). The most frequent infection was pneumonia (10/40; 25%) and the most common activity was renal (21/40; 53%). To diagnose”fever due to infection with-or-without activity”, compared to “fever only due to activity”, CRP had an AUC: .86 (95% CI: .75-.97) with 5.4 mg/dl as the best cut-off value (Se: 76.9%; Sp: 85.7%; PPV: 90.9; NPV: 66.6%). To diagnose “fever due to activity only”, compared to “fever due to infection with or without activity”, the ESR/CRP ratio had an AUC: .83 (95% CI: .68-.98) with 21.42 as the best cut-off value (Se: 78.6%; Sp: 84.6%; PPV: 73.3%; NPV: 88.0%). Conclusion: CRP and ESR/CRP ratio, which are routine tests in these cases, are potentially useful in discerning causes of fever in SLE patients. These findings should be confirmed in future studies.