Abstract
OBJECTIVE: To examine whether serum procalcitonin (PCT) concentrations are useful for distinguishing bacterial infections from disease flares in patients with systemic autoimmune diseases. METHODS: Patients with systemic autoimmune diseases who were admitted to our hospitals due to either a suspected deterioration of their primary diseases or an infectious disease were enrolled. Serum PCT levels were measured in 99 serum samples of 98 patients who had elevated serum C-reactive protein (CRP) levels; 29 samples were obtained from patients with bacterial infections, and 70 samples were obtained from patients with disease deterioration without a detectable infection. The diagnostic accuracy, sensitivity, and specificity for identifying a bacterial infection were estimated using the receiver-operating characteristic curve. Multiple logistic regression analysis was also done with PCT level, age, sex, steroid dose, and use of immunosuppressive agents. RESULTS: Serum PCT levels were higher in the bacterial infection group than in the disease flare group (mean +/- SD, 4.539 +/- 9.677 vs 0.116 +/- 0.127; p < 0.0001). The diagnostic accuracy of PCT for bacterial infection was 0.797, sensitivity 53.3%, and specificity 97.1%. On multivariate analysis, the odds ratio of a PCT > or = 0.5 ng/ml was significant (OR 59.085, 95% CI 7.705 453.088, p < 0.0001) for identifying bacterial infection. CONCLUSION: Elevated serum PCT levels have a good specificity for diagnosing bacterial infection in patients with systemic autoimmune diseases regardless of their dosage of oral corticosteroids and immunosuppressive agents.
