Abstract
Background: Although an etiological diagnosis is important for therapeutic decisions in children with pneumonia, this is often based on clinical, laboratory and radiological findings on admission.
Aim: To design a score for assessing etiology in children with pneumonia.
Subjects and methods: 175 children were studied, aged 1 to 59 months, hospitalized for pneumonia with its etiological diagnosis confirmed (127 viral and 48 bacterial). Predictor variables were age, axillar temperature, WBC, total neutrophils, band forms and chest X-ray (according to Khamapirad score). t-test or chi2 were used, when appropriate, to assess the associations between each predictor and the etiology. To make these variables dicotomic (except for radiology), the best cut-off value was established for each of them by receiver operating characteristic (ROC) curves. Logistic regression was used to obtain coefficients. By applying them and the X-ray value (according to Khamapirad) a score was defined. Finally, and using a ROC curve a cut-off value for predicting etiology (bacterial or viral) score was established.
Results: All predictors showed significant associations with outcome (p < 0.001). After logistic regression, the final score included: age >= 9 months (3 points), axillar temperature >= 39.0 °C (3 points), total neutrophils >= 8000 (3 points), band forms >= 5 % (1 point) and chest X-ray (-3 to 7 points). The cutoff value was established at >= 5 points where the score achieved a sensitivity of 97.9 %, a specificity 93.7 %, and a positive predictive value of 85.4 %, in the prediction of a bacterial etiology.
Conclusions: our score was useful in the assessment of a bacterial etiology in children with pneumonia. This core will allow faster and more accurate therapeutic decisions, allowing a more rational antibiotic use.
Article PDF
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Moreno, L., Duran, P., Bujedo, E. et al. 10 Designing a Score for Assesing Etiology in Children with Pneumonia. Pediatr Res 57, 921 (2005). https://doi.org/10.1203/00006450-200506000-00038
Issue Date:
DOI: https://doi.org/10.1203/00006450-200506000-00038