Background: Lower respiratory tract infection is the most common infection leading to unnecessary antibiotic treatment in children. Etiologic diagnosis is not immediately achieved, and the pathogen remains unidentified in a large number of cases. Neither clinical nor laboratory factors allow for a rapid distinction between bacterial and viral etiology. The aim of our study was to evaluate the reliability of procalcitonin (PCT), C-reactive protein (CRP) and leukocyte count in distinguishing pneumococcal, atypical and viral lower respiratory tract infection.
Methods: PCT, CRP and leukocyte count were measured in children with microbiologically documented diagnoses of lower respiratory tract infection. The results were compared of children with pneumococcal, atypical and viral etiologies.
Results: PCT and CRP showed significant correlation with a bacterial etiology of lower respiratory tract infection. No significance was found for leukocyte count. Using a cutoff point of 2 ng/ml for PCT and 65 mg/l for CRP, the sensitivities and specificities for distinguishing bacterial from viral lower respiratory tract infections were 68.6 and 79.4% for PCT and 79.1 and 67.1% for CRP. The sensitivities and specificities for distinguishing pneumococcal from other etiologies were 90.3 and 74.1% for PCT and 90.3 and 60% for CRP, respectively.
Conclusions: High PCT and CRP values show a significant correlation with the bacterial etiology of lower respiratory tract infection. PCT and CRP show good sensitivity for distinguishing pneumococcal from other etiologies. PCT shows higher specificity than CRP. PCT and CRP can help make decisions about antibiotic therapy in children with lower respiratory tract infections.