- SBI —
- serious bacterial infection
- UCS —
- unstructured clinical suspicion
- YOS —
- Yale observation scale
Febrile infants <3 months of age, especially those <1 month, are at risk for serious bacterial infection (SBI), urinary tract infections, bacteremia, and meningitis. In this issue of Pediatrics, Nigrovic et al1 report on 5997 infants ≤60 days of age with fever ≥38.0°C enrolled in 14 pediatric emergency departments over a 4.5-year period in the Pediatric Emergency Care Applied Research Network. The analysis focuses on the performance of the Yale observation scale (YOS) and unstructured clinical suspicion (UCS) in detecting the 444 infants with SBIs, 97 of whom had invasive bacterial illnesses, meningitis, or bacteremia. Neither form of assessment was sufficiently sensitive in detecting SBIs. For example, of the 24 infants with bacterial meningitis, only 10 had a high YOS score of >10; of the 23 infants with meningitis assessed by UCS, only 11 had high scores.
The YOS was developed to capture the instinctive clinical judgments that pediatricians make through observation to assess a child’s degree of illness.2 In analyzing 186 domains (eg, motor behavior) of observation data used to formulate these scales, ∼60% focused on the child’s response to stimuli.3 To accurately assess the degree of illness through observation, the examining pediatrician is a clinician and a developmentalist who observes abnormal physical findings (eg, rash), provides age-appropriate stimuli, and assesses age-appropriate responses. Development of the YOS focused on detecting serious illness in febrile children 0 to 36 months of age; 3 of the 6 items of the YOS focus on stimulus-response data. Infants in the first 2 to 3 months of life have a less developed repertoire of responses to stimuli than do older children, and when the diagnostic efficacy of the YOS in febrile infants was tested in a previous study, their sensitivity was suboptimal.4 UCS faces the same challenge. Nonetheless, Baker et al5 include assessment of appearance by the YOS as a component of the physical examination and laboratory testing diagnostic approach for febrile infants who were 29 to 56 days of age, the sensitivity of which for bacteremia and meningitis was 96%, detecting 27 of 28 affected infants.
Other studies have documented the importance of observational assessment of febrile infants. The Pediatric Research in Office Settings study involved 573 practitioners in 44 states.6 Infants 0 to 3 months of age with fever had a standard observational assessment based on domains in the YOS. In a 3-year period 3066 infants were enrolled. There were 63 cases of bacteremia and/or bacterial meningitis, with greater occurrence in the first month of life. The observational assessment of moderately or very ill (corresponding to YOS scores of 11–15 and ≥16) detected 36 (58%) of the 63 infants and was considered an important component of a diagnostic approach, which included appearance, age <25 days, fever >38.6°C, and had a sensitivity for bacteremia and/or bacterial meningitis of 94% (59 of 63 infants). Greenhow et al7 reported on 1380 infants who were 7 to 90 days of age with fever seen in Northern California Kaiser Permanente over a 3-year period, 40 of whom had bacteremia and/or meningitis. Laboratory testing was variable in frequency but more common in infants 7 to 28 days old. Appearance was a determinant of testing. In ill-appearing infants 7 to 28 days old, the occurrence of SBI was high (33%). There was no delayed identification of bacteremia and/or meningitis in the entire cohort. Hence, these 2 studies are more supportive of the efficacy of observational assessment in evaluating febrile infants.
Nigrovic et al1 recognize some limitations in their report. There was no testing of reliability in application of the YOS or of the nonvalidated scale for UCS. Perhaps most important is the exclusion of infants who were critically ill with signs of sepsis. The authors did not estimate the number of these patients. The YOS was developed to identify ill children, and the exclusion of infants critically ill in the Pediatric Emergency Care Applied Research Network study likely reduced the sensitivity of both the YOS and UCS. In the Pediatric Research in Office Settings study, such infants were not excluded.
The cited studies support the following:
The YOS and UCS have limitations in their diagnostic efficacy in evaluating febrile infants likely related to the limited repertoire in infant responses to stimuli on which such assessments are based.
Nonetheless, there is an important place as reported by Baker et al,5 Pantell and co-workers,7 and Greenhow et al,7 for the assessment of degree of illness by observation in the diagnostic approach to febrile infants.
Observational assessment should continue to be an important component of the diagnostic assessment of febrile infants, but a broad approach of age assessment, careful history and physical examination, and laboratory testing is required to detect all seriously ill febrile infants. Current paradigms of this broad approach have been reported5,6 but do not achieve 100% sensitivity for this detection. Hence, I agree with the conclusion of Nigrovic et al,1 namely that research to identify laboratory parameters that will increase sensitivity of the diagnostic approach is warranted.
- Accepted April 20, 2017.
- Address correspondence to Paul L. McCarthy, MD, Department of Pediatrics, Yale School of Medicine, 333 Cedar St, New Haven, CT 06520. E-mail:
Opinions expressed in these commentaries are those of the author and not necessarily those of the American Academy of Pediatrics or its Committees.
FINANCIAL DISCLOSURE: The author has indicated he has no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The author has indicated he has no potential conflicts of interest to disclose.
COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2017-0695.
- Nigrovic LE,
- Mahajan PV,
- Blumberg SM, et al; Febrile Infant Working Group of the Pediatric Emergency Care Applied Research Network (PECARN)
- McCarthy PL,
- Sharpe MR,
- Spiesel SZ, et al
- McCarthy PL,
- Jekel JF,
- Stashwick CA, et al
- Baker MD,
- Avner JR,
- Bell LM
- Greenhow TL,
- Hung YY,
- Pantell RH
- Copyright © 2017 by the American Academy of Pediatrics