Do Febrile Infants Aged 60 to 90 Days With Bronchiolitis Require a Septic Evaluation?

Abstract

Febrile infants younger than 90 days and with bronchiolitis have very low rates of serious bacterial infection. The authors searched the National Library of Medicine Medline database (to December 2010). In addition, the bibliographies of these studies were searched for studies not otherwise identified. Studies were selected if they reported the incidence of site-specific, concomitant serious bacterial infection in the setting of fever and clinical bronchiolitis or documented respiratory syncytial virus infection. Studies also had to report the type of culture collected and contain age-specific data for infants younger than 60 to 90 days. The authors extracted the event rates for bacteremia, meningitis, and urinary tract infection. Urinary tract infections were diagnosed by urine culture. Two authors reviewed and graded each study for the level of evidence. Grading system was not described. All 3 authors resolved discrepancies by consensus. A random-effects meta-analysis of urinary tract infection was performed. Tabled 1Rate of urinary tract infection in infants younger than 90 days with bronchiolitis.Method to Diagnose BronchiolitisNumber of StudiesRate (95% Confidence Interval)Clinical diagnosis62.0% (1.2–3.5)Respiratory syncytial virus testing55.1% (2.6–9.6)All studies113.3% (1.9–5.7) Open table in a new tab The study team identified 118 relevant articles, of which 11 met all inclusion criteria and were analyzed. Six studies had a retrospective design. The diagnosis of bronchiolitis was made by clinical symptoms in 6 studies and by respiratory syncytial virus testing in 5. Subjects younger than 90 days were included in 7 studies and younger than 60 days, in 4 studies. There was a significant variability in the clinical settings (emergency department, inpatient, outpatient, and mixed). Significant statistical heterogeneity was attributed to the inclusion criteria, not to study setting or design. Results of heterogeneity testing were not reported. The authors performed a meta-analysis only for urinary tract infection and bronchiolitis (Table). Results were not pooled for bacteremia and meningitis because of the very low event rates reported. When evaluating infants with symptoms consistent with bronchiolitis, physicians are often concerned with ensuring the lack of more serious bacterial infections, including urinary tract infection, occult bacteremia, and meningitis. As a result, many of these infants, particularly those younger than 60 to 90 days, are screened for occult bacterial infections with blood cultures, urine culture, and spinal fluid analysis.1Craig C.J. Williams G.J. Jones M. et al.The accuracy of clinical symptoms and signs for diagnosis of serious bacterial infection in young febrile children: prospective cohort study of 15,781 febrile illnesses.BMJ. 2010; 340: c1594Crossref PubMed Scopus (221) Google Scholar For infants who present with typical bronchiolitis symptoms, lumbar punctures and blood cultures are unnecessary. Routine blood cultures show a less than 1% rate of occult bacteremia, whereas false-positive blood culture rates range from 0.9% to 3.6%.2Joffe M.D. Alpern E.R. Occult pneumococcal bacteremia: a review.Pediatr Emerg Care. 2010; 26: 448-457Crossref PubMed Scopus (32) Google Scholar These recommendations cannot be applied to toxic-appearing infants. In contrast, there is a 3.3% rate of concomitant urinary tract infection in patients receiving a diagnosis of bronchiolitis. A large prospective study of febrile children reported a similar urinary tract infection rate.1Craig C.J. Williams G.J. Jones M. et al.The accuracy of clinical symptoms and signs for diagnosis of serious bacterial infection in young febrile children: prospective cohort study of 15,781 febrile illnesses.BMJ. 2010; 340: c1594Crossref PubMed Scopus (221) Google Scholar This review was unable to rule out the possibility of asymptomatic bacteriuria (>100,000 colony-forming units in the urine, without urinary tract infection symptoms) primarily because of patient age. Previous study has shown the rate of asymptomatic bacteriuria to be 2.5% for boys and 0.9% for girls in the first year of life.3Wettergren B. Jodal U. Spontaneous clearance of asymptomatic bacteriuria in infants.Acta Paediatr Scand. 1990; 79: 300-304Crossref PubMed Scopus (43) Google Scholar Clinical judgment should be exercised to determine whether additional testing is warranted for these children.