Is a chest radiograph necessary in the evaluation of every febrile infant less than 8 weeks of age?

Pediatrics. 1991 Oct;88(4):821-4.

Abstract

This study was designed to examine the relationship between respiratory signs and the likelihood of having an abnormal chest radiograph in a sample of febrile infants less than 8 weeks of age. The sample consisted of 242 infants who were admitted during a 3-year period with temperatures greater than or equal to 38 degrees C (100.4 degrees F) and had a chest radiograph. The house officer recorded the presence of respiratory signs and symptoms including rhinorrhea, tachypnea, cough, rales, wheezes, retractions, and rhonchi. Each chest radiograph was reviewed independently according to predetermined criteria by a senior radiology resident and an attending pediatric radiologist. Interobserver agreement was 91%. Both observers were blind to the infants' respiratory signs. The chest radiograph interpretations were compared with the presence of respiratory signs. Of the 242 cases, 228 had chest radiographs available for interpretation. Of these, 27 chest radiographs (12%) were identified as abnormal, including 6 where there was initial disagreement as to the presence of an abnormality. Twenty-five (31%) of 80 infants with any respiratory signs had an abnormal chest radiograph, whereas only 2 (1%) of 148 asymptomatic infants did. The sensitivity of respiratory signs was 93% (confidence interval = 76% to 99%). These findings suggest that in the absence of respiratory signs, febrile infants are unlikely to have an abnormal chest radiograph.

MeSH terms

  • Fever / diagnosis
  • Fever / etiology*
  • Humans
  • Infant
  • Observer Variation
  • Radiography, Thoracic*
  • Respiratory Tract Diseases / complications
  • Respiratory Tract Diseases / diagnostic imaging*
  • Sensitivity and Specificity