Today’s Pediatric grand rounds was given by Dr. Prashant Mahajan, MH, MPH, MBA.
Professor and Vice-chair of the department of Emergency Medicine; Professor of Pediatric Medicine, Division Chief of Pediatric Emergency Medicine, Professor of Pediatrics at the University of Michigan
For those that don’t know him- He’s really smart and has done a ton of research on febrile infants
and he's proposing a new model to rule out serious bacterial infections in infants <60 days old.
TL:DR
Serious Bacterial Infection (SBI) can be ruled out febrile infants from 29-60 days old with a-
Negative UA
Absolute Neutrophil Count (ANC) < 4090/μg
Procalcitonin < 1.72 ng/m
This prediction rule has a
Sensitivity of 97.7% (95% CI, 91.3-99.6)
Negative predictive value of 99.6% (95% CI, 98.4-99.9)
Negative likelihood ratio of 0.04 (95% CI, 0.01-0.15)
Specificity of 60.0% (95% CI, 56.6-63.3)
This rule requires further validation, but has promise to substantially decrease the use of lumbar punctures, broad-spectrum antibiotics, and hospitalization for many febrile infants 60 days and younger.
The longer and more detailed approach:
- 8-13% of infants <60 days with fever have a SBI
~5-8% have a UTI
~1-2% have bacteremia
~0.5% have meningitis
- ~500,000 febrile infants are evaluated by healthcare professionals annually
Missed SBIs may lead to serious complications
Febrile infants frequently receive invasive management including lumbar punctures, broad spectrum antibiotics, and hospitalization
Variation exists in the management of febrile infants <60 days
90% of those 28 days or less receive lumbar puncture and admission
The incidence of SBIs has decreased over time
We need to balance hospital related complications, costs, and increases in antimicrobial resistance with the consequences of missed SBIs
- Our screening tests to assess for SBIs have holes in them
Physical Exam
Yale Observation Scores (YOS) in infants with SBI’s have similar median scores to those without SBI’s
I didn't know the YOS was a thing either. It's a clinical score developed on 6 behavioral domains to predict SBI’s,
CBC’s are not sensitive in ruling out bacteremia or meningitis
WBC< 5,000 has a sensitivity of 10%, specificity of 91%
WBC> 15,000 has a sensitivity of 18%, specificity of 87%
Several of the commonly used rules for febrile infants (Philadelphia, Rochester, Boston, and Pittsburgh) were not statistically derived and therefore lacked optimal balance between test sensitivity (avoiding missed SBIs) and specificity (preventing overtesting and overtreating patients without SBIs). Additionally, several included data from LP’s an invasive procedure not required in the newly proposed rule (Boston, Phladelphia, Pittsburgh, Milwaukee.
- In this study
Negative UA alone ruled out an SBI in 97.6% of cases
Anyone hear of diapedesis? Consult Hector Vazquez for more info
Negative UA + ANC <4090 ruled out SBI in 99.2% of cases
Negative UA + ANC <4090 + PCT <1.71 ruled out SBI in 99.8% of cases
- Further validation in a cohort with more SBI’s is needed before implementation of this new rule.
The conclusion
Dr. Mahajan recommends using this rule in infants 29-60 days old. He currently recommends pursing your institutions’s standard of care (Full Sepsis Workup) in infants 28 days old or less
The Article:
https://jamanetwork.com/journals/jamapediatrics/fullarticle/2725042