• C A Scirè, L Cavagna, C Perotti, E Bruschi, R Caporali, and C Montecucco.
• Rheumatology and Rheumatology Unit, University of Pavia, IRCCS Policlinico S. Matteo and Immunohematology Unit, IRCCS Policlinico S. Matteo, Pavia, Italy.
• Clin Exp Rheumatol. 2006 Mar 1;24(2):123-8.

ObjectiveTo determine the usefulness of plasma procalcitonin (PCT) measurement to suspect infectious etiology in febrile patients with systemic autoimmune disease.MethodsPCT, C-Reactive protein (CRP), erythrocyte sedimentation rate (ESR) and white blood cell count (WBC) were measured in 44 consecutive inpatients with a diagnosis of systemic autoimmune disease and fever >38 masculine C. After careful microbiologic screening no obvious infection was demonstrated in 24 patients (Group A) while an infectious bacterial complication was diagnosed in 20 cases (Group B).ResultsMedian PCT levels were significantly higher in the group B (1.11 vs 0.24 ng/ml; p = 0.0007), whereas the differences for CRP, WBC and ESR did not reach statistical significance. PCT also exhibited a good sensitivity and specificity (75%) in differentiating patients with infection from those with disease flare. With respect to positive and negative predictive values (71.4% and 78.2%), PCT markedly exceeded the other variables. By analyzing PCT values by disease we identified a false positive subgroup of patients suffering from adult onset Still's disease (AOSD), showing markedly elevated PCT levels in absence of infection. By excluding these patients, PCT showed a very good sensitivity and specificity (73.6% and 89.4%) and the area under receiver operating characteristics (ROC) curve rose from 0.801 to 0.904.ConclusionOur data indicate that elevated PCT concentrations offer good sensitivity and specificity for the diagnosis of systemic bacterial infection in febrile patients with systemic autoimmune diseases. However, in fever associated with AOSD PCT may be elevated even in the absence of infectious complication.

26 keywords...

hide…