Lack of predictive value of tachypnea in the diagnosis of pneumonia in children.

BACKGROUND: The World Health Organization (WHO) recommends the use of tachypnea as a proxy to the diagnosis of pneumonia in resource poor settings.
OBJECTIVE: To assess the relation between tachypnea and radiographic pneumonia among children evaluated in a pediatric emergency department (ED).
METHODS: Prospective study of children less than 5 years of age undergoing chest radiography (CXR) for possible pneumonia was conducted in an academic pediatric ED. Tachypnea was defined using 3 different measurements: (1) mean triage respiratory rate (RR) by age group, (2) age-defined tachypnea based on WHO guidelines (<2 months [RR > or =60/min], 2 to 12 months [RR > or =50], 1 to 5 years [RR > or =40]), and (3) physician-assessed tachypnea based on clinical impression assessed before CXR. The presence of pneumonia on CXR was determined by an attending radiologist.
RESULTS: A total of 1622 patients were studied, of whom, 235 (14.5%) had radiographic pneumonia. Mean triage RR among children with pneumonia (RR = 39/min) did not differ from children without pneumonia (RR = 38/min). Twenty percent of children with tachypnea as defined by WHO age-specific cut-points had pneumonia, compared with 12% of children without tachypnea (P < 0.001). Seventeen percent of children who were assessed to be tachypneic by the treating physician had pneumonia, compared with 13% of children without tachypnea (P = 0.07).
CONCLUSION: Among an ED population of children who have a CXR performed to assess for pneumonia, RR alone, and subjective clinical impression of tachypnea did not discriminate children with and without radiographic pneumonia. However, children with tachypnea as defined by WHO RR thresholds were more likely to have pneumonia than children without tachypnea.