FEVER! It demands a lot of our attention when caring for children. Appropriately, it has also been baked into many delicious Morsels. Certainly, there are numerous Infectious Diseases to consider and, of course, there are concerns when we can’t discern an obvious origin of the fever. Then of course, we have all learned from 2020 to never say “it’s just a virus” ever again. Yet, the one thing that constantly generates more “facepalms” and “eye-rolls” is the evolving landscape of how to appropriately evaluate the infant with a fever. (Do we apply Rochester criteria or Philadelphia criteria? What about Boston? Is there another city we should consider? What if I don’t like the northeastern portions of the US? At least we agree that bagged urines are terrible.) In an effort to better know the landscape and what is on the horizon, let us review another city with a name in the fever game – Roseville Criteria for Pediatric Fever Evaluation:

Pediatric Fever: < 60 day olds

Neonates and Infants like to hide things.

The toxic-appearing neonate or infant may not tell you what is wrong initially, but you are going to DO EVERYTHING!
- The overtly ill infant clearly needs not only a full septic work-up and empiric antibiotics and antivirals (don’t forget HSV), but also considerations for a broad Ddx.
- Abridged Ddx: CAH, Inborn-Errors of Metabolism, Trauma, Coarctation, Congenital Heart Disease.

The well-appearing, febrile neonate and infant (especially < 60 days of age), on the other hand, is quite good at deceiving us into believing that they are only dealing with a minor illness.
- At baseline, they don’t do much… so an illness may not initially cause them to behave drastically different.
- Our ability to find subtle clues of significant illness is also limited since their physical exams may not be dramatically altered.
  - Great reasons to be vigilant about checking capillary refill and muscle tone/posture.
  - Knowing milestones may help as well.
- At this age, they are known to hide (ie, have occult):
  - Meningitis
  - Bacteremia
  - Pneumonia
  - UTI
  - Osteomyelitis

The question has long been – how do you best detect these occult conditions without doing more harm than good (ex, false positive cultures, invasive testing, hosptializations)?
- The answer… well… is still debated.
- Fortunately, vaccinations have helped tremendously for the older infants (> 60 days… since you get 1st set at 2 months).
- Screening criteria have been widely used over the past several decades, but information continues to evolve.

UTI and Concurrent Meningitis

Way back in 2012, we discussed the conundrum of encountering an abnormal urinalysis result while screening a febrile infant.
- If you are using the Rochester Criteria (for example), then that infant would require a lumbar puncture.
- Yet, there was growing evidence that the rates of concurrent meningitis and UTI in infants is low… so, is that LP exposing the child to more risk than reward?
  - Neonates still have high rates of meningitis concurrently with UTI (NEVER TRUST A NEONATE).
  - > 29 day olds: the collective group has been shown to have a low risk for concurrent meningitis with UTI – between 0% and 0.74% (so, very low).
So does one follow the guidelines or the evolving evidence? Perhaps… now you can do both… with the Roseville Criteria.

Roseville Criteria for Pediatric Fever Evaluation

Roseville Criteria [PMID 33334815: Nguyen, 2021]

Evaluation is contingent upon Examination, CBC with Differential, Urinalysis and Cultures (Blood and Urine… and maybe CSF).

Is based on Rochester criteria, but:
- Considers neonates (7-28 days) and infants (29-6 days) separately
- Adds additional High Risk Criterion for neonates – Temperature > 38.5 degrees C.
- Allows for definition of “low risk” neonates (7-28 days) and for their management without LP and outpatient close follow-up.
- Allows for management of infant with only high risk criteria being Urinalysis to be managed with oral antibiotics and as an outpatient.

Neonates (7-28 days)
- Low Risk Criteria for defined as: [PMID 33334815: Nguyen, 2021]
  - Temperature </= 38.4 degrees C (101.2 degrees F)
  - WBC > 5,000 or < 15,000
  - Absolute Band Count < 1500
  - Urinalysis with < 5 WBC and NO Leukocyte Esterase or Nitrites
- High risk neonates should get LP and IV antibiotics.
- One caveat: neonate who is high risk solely due to temperature > 38.5 degrees C and has a NORMAL CSF analysis can be observed in hospital off of antibiotics.
- Low Risk neonates (according this this protocol) can be managed without lumbar puncture and discharged home.
  - PERSONAL DISCLAIMER:
  - I may be too old for this approach… I still, personally favor the LP for this age group.
  - If I am unable to get the CSF, and the child meets all LOW Risk criteria, then admission with no antibiotics is at least prudent with this guideline.
  - I will keep an open mind and try to evolve though… .. .

Infants (29-60 days)
- Low Risk Criteria for defined as: [PMID 33334815: Nguyen, 2021]
  - WBC > 5,000 or < 15,000
  - Absolute Band Count < 1500
  - Urinalysis with < 5 WBC and NO Leukocyte Esterase or Nitrites
- Infants with High Risk WBC and/or Band Count should get parenteral antibiotics and a lumbar puncture should be considered.
- Infants with ONLY High Risk Urinalysis can be given ORAL antibiotics and discharged home with close follow-up.

Excluded infants: [PMID 33334815: Nguyen, 2021]
- Preterm (<37 weeks GA)
- Ill appearing (see first bullet above)
- < 7 days of age
- History of perinatal maternal or infant fever or antibiotic treatment
- Congenital anomalies
- Technology dependence
- Evident Infection (ex, abscess, cellulitis, bacterial conjunctivitis, mastitis, omphalitis, perforated otitis media)

Roseville Criteria Performance [PMID 33334815: Nguyen, 2021]

Nguyen et al. looked performance of the Roseville protocol by retrospectively examining cases to include 627 neonates and 1176 infants.
- For Neonates: Sensitivity = 96.7%; Negative Predictive Value = 99.5%
- For Infants: Sensitivity = 91.4%; Negative Predictive Value = 99.6%

It performed favorably compared to Rochester, Philadelphia, and Boston criteria.
- Discharging well-appearing febrile infants (29-60 days) who had abnormal urinalysis only on oral antibiotics would reduce IV antibiotics (hospitalizations) and lumbar punctures.
- This reduction was achieved while retaining the sensitivity and NPV compared to other protocols.
- No increase in missed cases of meningitis.

Some things to consider:
- Exclusion matters. Do not apply these protocols to patients who the protocol specifically excluded!
- Meningitis is RARE(<1%) and, thus, it can be difficult to draw definitive conclusions about the ability of the studied protocol to reliably exclude it.
- Bacteremia is important to consider… but of those with presumptive UTI who ended up with bacteremia concurrently none had meningitis.
- Retrospective studies have inherent weaknesses… but a prospective double-blind case-controlled study is difficult to perform in children.
- Research will continue to evolve:
  - PECARN has a prediction rule (although it includes LP for patients with UTI). [PMID 30776077: Kuppermann, 2019]
  - Data on new biomarkers will continue to require reconsideration of these protocols.
- Outpatient plan matters:
  - In order to send an infant home on oral antibiotics for a presumptive UTI, ensure the PCP is on the same page and make sure the family has appropriate education and resources.
  - The child and the family need to prove to you that they are suitable for this newer version of the fever management pathway.

Moral of the Morsel

Science evolves as data evolves. It is part of our responsibility to attempt to stay current.
Practice patterns should not be stagnant. They should also not vary based on the direction of the wind.
Save the LP when appropriate. UTI can be managed in a 29-60 day infant as an outpatient if you apply the Roseville criteria.
Don’t forget to ensure all of the “i’s” are dotted and “t’s” crossed. Double check the outpatient plan is acceptable by ALL parties involved.