Procalcitonin kinetics: a reliable tool for diagnosis and monitoring of the course of bacterial infection in critically ill patients with autoimmune diseases

Abstract

Dear Editor, Early diagnosis of infection in patients with autoimmune diseases is important but can be challenging [1, 2]. Procalcitonin (PCT) has been proposed as a biomarker of bacterial infection. However, its role is still disputed and has not been assessed in critically ill patients with autoimmune diseases, especially its kinetics [3, 4]. We determined the value of PCT kinetics for monitoring the cause of sepsis in this population. After approval by the ethics committee of our hospital, 81 autoimmune disease patients with systemic inflammatory response syndrome (SIRS) admitted to intensive care unit (ICU) were screened between January 2008 and December 2010. Sixty were enrolled (70.7 ± 12.7 years, 38 females) and grouped as bacterial infection (n = 31) or without bacterial infection (n = 29) (Fig. 1). The Acute Physiology and Chronic Health Evaluation (APACHE) II score was 17.3 ± 8.9, and 28-day ICU crude mortality was 38.3 %. PCT was collected on day 1 and day 2, and at 2 weeks or on the day of discharge from ICU (day-e) and determined by immunoluminometric assay (BRAHMS, Berlin, Germany). The serum PCT level [median (range), lg/L] was 1.53 (0.08–32.5) in all patients on day 1, being significantly higher in patients with bacterial infections than in other patients [1.88 (0.12–32.5) versus 0.57 (0.08–7.83), p = 0.002], but with no difference between subgroups of nonbacterial infection (n = 17) and disease flare (n = 12) [0.61 (0.08–7.83) versus 0.55 (0.10–7.78), p = 0.472] (Fig. 2). In receiver operating characteristic (ROC) curve analysis, PCT on day 1 was the earliest predictor for bacterial infection [area under curve (AUC) = 0.787; 95 % confidence interval (CI): 0.615–0.929; p = 0.006], and the threshold of 1.05 lg/L had sensitivity of 74.2 %, specificity of 86.2 %, positive likelihood ratio of 5.37, and negative likelihood ratio of 0.29. Furthermore, PCT levels for survivors (n = 37) did not differ significantly from nonsurvivors (n = 23) on day 1 [1.51 (0.08–32.5) versus 1.55 (0.08–29.8), p = 0.537] or day 2 [1.59 (0.10–105.0) versus 1.57 (0.10–110.5) lg/L, p = 0.479], but were significantly higher in nonsurvivors on day-e [5.12 (0.87–110) versus 0.23 (0.08–2.87) lg/L, p = 0.001] and had a strong correlation with outcome (r = 0.66, p = 0.007). On the contrary, PCT declined gradually in survivors with or without bacterial infection from day 1 to day-e [bacterial infection