Background: The aim of this study was to differentiate unspecific and self-limiting fever after bronchoscopy from fever due to infection by using serum procalcitonin, C-reactive protein and neutrophil count. Furthermore, frequency of fever after bronchoscopy and procedures as possible risk factors were evaluated.
Methods: Three hundred and fourteen consecutive patients were included. All bronchoscopies were performed using jet-ventilation and general anesthesia. Patients were analyzed according to interventions performed during bronchoscopy and laboratory results. Microbiological assessment was done in patients who developed fever to prove or rule out a bacterial infection.
Results: Forty-four patients showed fever within 24 h following bronchoscopy (14%). A bacterial infection was proven in 11 patients with fever (3.5%). Procalcitonin, neutrophil count and C-reactive protein were significantly higher in patients with fever after bronchoscopy compared to non-fever patients. To predict bacterial infection in the receiver operating analysis, procalcitonin had the highest area under the curve (0.942; 95% confidence interval [CI], 0.768 to 1.000; p = <0.001), followed by neutrophil count (AUC, 0.804; 95% CI, 0.606 to 0.946; p = 0.005), whereas CRP levels where not statistically significant. Endoscopic airway recanalization was the only intervention that induced fever more frequently than all other interventions (OR 13.629).
Conclusions: Fever is frequently seen after bronchoscopy and in some cases caused by bacterial infection. Procalcitonin might be useful to distinguish a bacterial infection from unspecific self-limiting fever. Airway recanalization is a procedure that seems to induce fever significantly more often than other bronchoscopic interventions.