Related Experiment Video
Updated: Jun 21, 2026
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
Inflammatory parameters predict etiologic patterns but do not allow for individual prediction of etiology in patients
Stefan Krüger1, Santiago Ewig, Jana Papassotiriou
1Medical Clinic I, University Clinic RWTH Aachen, Germany. stkrueger@ukaachen.de
Background:
Aim of this study was to evaluate the correlation of inflammatory markers procalcitonin (PCT), C-reactive protein (CRP) and leukocyte count (WBC) with microbiological etiology of CAP.
Methods:
We enrolled 1337 patients (62 +/- 18 y, 45% f) with proven CAP. Extensive microbiological workup was performed. In all patients PCT, CRP, WBC and CRB-65 score were determined. Patients were classified according to microbial diagnosis and CRB-65 score.
Results:
In patients with typical bacterial CAP, levels of PCT, CRP and WBC were significantly higher compared to CAP of atypical or viral etiology. There were no significant differences in PCT, CRP and WBC in patients with atypical or viral etiology of CAP. In contrast to CRP and WBC, PCT markedly increased with severity of CAP as measured by CRB-65 score (p < 0.0001). In ROC analysis for discrimination of patients with CRB-65 scores > 1, AUC for PCT was 0.69 (95% CI 0.66 to 0.71), which was higher compared to CRP and WBC (p < 0.0001). CRB-65, PCT, CRP and WBC were higher (p < 0.0001) in hospitalised patients in comparison to outpatients.
Conclusion:
PCT, CRP and WBC are highest in typical bacterial etiology in CAP but do not allow individual prediction of etiology. In contrast to CRP and WBC, PCT is useful in severity assessment of CAP.
