- Poster presentation
- Open Access
Role of procalcitonin as a diagnostic and prognostic marker of infectious diseases in the emergency department: preliminary data
© BioMed Central Ltd. 2010
- Published: 1 March 2010
- Emergency Department
- Antibiotic Therapy
- Prognostic Marker
- Septic Patient
- Significant Statistical Difference
An observational study was conducted in a teaching hospital to evaluate the diagnostic and prognostic value of procalcitonin together with C-reactive protein (CRP) in assessing the presence of infection/sepsis in patients arriving in the emergency department (ED). The objective was to improve diagnostic and therapeutic time to intervention by the emergency physician.
Three hundred and five patients were studied (145 male, 160 female, mean age 67.2 years). Vital parameters were recorded during patient's stay in the ED. Hemocultures and biological fluids cultures were performed before antibiotic therapy. CRP and procalcitonin were performed at arrival in the ED, and after 5 days of antibiotic therapy.
Median PCT and CRP values at arrival in the ED were 0.3 ng/ml and 12.9 mg/dl, after 5 days they were 0.09 ng/ml and 4.2 mg/dl with a significant statistical difference in both markers (P < 0.001). PCT was higher in sepsis vs infections (respiratory, urinary, gut or skin infections) (3.1 vs 0.2 ng/ml, P < 0.001). PCT pre-therapy was significantly higher in septic nonsurvivors vs septic survivors (1.0 vs 0.2 ng/ml, P < 0.001), and there was a significant PCT increase after 5 days in nonsurvivors vs survivors (3.6 vs 0.07 ng/ml) (P < 0.001). CRP results did not show a statistical significance, except for CRP measured after 5 days (3 mg/dl) vs CRP pre-therapy in survivors (12 mg/dl, P < 0.001). PCT (cut-off 0.5 ng/ml) ROC curve showed an AUC 0.72 (P < 0.001), with a sensitivity 75% as a diagnostic marker for infection. In septic patients, PCT ROC curve had an AUC 0.70 (P < 0.001) if considered as a prognostic factor of death (cut-off 1 ng/ml). Although pre-therapy PCT values were not particularly sensitive, nor specific for predicting deaths, PCT in therapy (cut-off 0.5 ng/ml) had an AUC 0.79 (sensitivity 75%, specificity 76% in predicting events) (P < 0.0001). This was higher than CRP (AUC 0.73, P < 0.0004).
PCT, with CRP, is useful to make an initial rapid infection diagnosis for patients presenting to the ED, and also to discriminate it from sepsis. Our results are in agreement with other authors [1, 2]. A PCT cut-off value of 0.5 ng/ml shows high sensitivity as a diagnostic infection marker, but also as a prognostic marker for outcomes. PCT with CRP are useful tools to assess a fast empiric therapy, and to ameliorate the prognosis of infectious/septic in ED patients.