Objective: To assess the value of procalcitonin (PCT) measurement to differentiate infection from non-infection in critically ill patients requiring long-term immunosuppressive therapy.
Methods: A prospective study was conducted in patients with underlying diseases requiring corticosteroids or chemotherapy in ICU from January 2008 to December 2009. Patients were divided into the infection group and the non-infection group and their PCT levels were compared.
Results: A total of 103 patients (65 women) were enrolled in this prospective study [aged (47.9 ± 21.9) years old] with 84 in the infection group and 19 in the non-infection group. The baseline level of PCT was significantly higher in infection than in non-infection patients [2.58 (0.08 - 44.65) pg/L vs 0.62 (0.15 - 6.00) pg/L, P = 0.002]. Different levels of PCT were manifested in different pathogen groups with 3.41 (0.45 - 44.65) pg/L in bacteria infection, 0.99 (0.28 - 6.67) pg/L in fungus infection, 0.11 (0.08 - 0.20) pg/L in virus infection group (P = 0.018). The AUC(ROC) of PCT was 0.867 for diagnostic bacterial infection. By multivariate analysis, the factors associated with the level of PCT were bacteria infection (OR 5.1, P = 0.031) and septic shock (OR 7.5, P = 0.027), while the factors not associated with the level of PCT were age, renal function, infection site and prognosis (P > 0.05).
Conclusions: The level of PCT is increased in the critically ill patients requiring immunosuppressive therapy with infection and it can be used for diagnosis for bacterial infection.