In this post I link to and excerpt from ‘The “48 Hour Rule-out” for Well-Appearing Febrile Infants’ from Episode #4 of the Cribsiders [Link is to the podcast and show notes].
However, I think that Episode #4 should be considered along with a careful review of Validation of the “Step-by-Step” Approach in the Management of Young Febrile Infants [PubMed Abstract] [Full Text HTML] [Full Text PDF]. Pediatrics August 2016, 138 (2) e20154381.
Here are the time stamps for the above portion of the podcast:
- The “48 hour rule-out” for the febrile infant 26:30
- When you should keep a child in the hospital for longer observation 24:00
- Overall recommendation for sepsis rule out in febrile infant 38:50
Here is the podcast:
Here are excerpts from The “48 Hour Rule-out” for Well-Appearing Febrile Infants:
History of the 48-hour rule-out
The 48 hour rule-out is based on studies that were performed in the 1970s, and there are two main reasons why they are not necessarily relevant anymore.
First, this was a time-period when culture plates were physically examined by a human to check for growth; therefore, they were only checked once a day (Rowley 1986), meaning that the time to positivity (TTP) was artificially long.
[Now machines monitor cultures for growth and alert lab tech as soon as the culture is positive.]
Second, these studies were overly broad and do not include our typical 48-hour rule-out patient, which is a previously healthy infant that has been discharged from the hospital. Their studies included NICU infants and ICU admissions that often grow atypical organisms, such as fungi, that need longer to grow out, which inflated the time to positivity of their cultures (Rowley 1986; La Scolea 1981; Pichichero 1979).
Current time to positivity of cultures
Recent studies of healthy febrile infants with new culture monitoring systems report that TTP for 97% of bacteria treated as true pathogens is ≤36 hours (Evans 2013). Additionally, after 36 hours, the rate of identifying a contaminant increases by 8-fold. No difference was found in infants ≤ 28 days versus those aged 0-90 days. Overall, the mean TTP in infants aged 0-90 days was 15.4 hours, and only 4% of possible pathogens were identified after 36 hours. In other words, you would have to monitor 1,250-2,778 infants past 36 hours in order to catch one bacteremic infant.
Up-To-Date has an excellent article, Febrile infant (younger than 90 days of age): Outpatient evaluation. “Literature review current through: Jul 2020. | This topic last updated: May 26, 2020.”
How to determine a low-risk infant & limitations
There are well documented clinical criteria via the Rochester (Jaskiewicz 1994), Boston (Baskin 1992), and Philadelphia (Baker 1993) studies that can be used to formally decide (Herzke 2018). Dr. Herzke considers monitoring an infant for longer than 36 hours if there is a clinical suspicion for HSV, abnormalities on CSF/urine studies, and/or if the child is not well-appearing.
Recommendations to Change Practice
- Use any of the above scores that you would prefer to determine if your infant is low risk
- If your infant fits the criteria and is well-appearing, you can feel comfortable discharging them at 24 to 36 hours
- If you have good follow-up, you can consider discharge as early as 24 hours
- Expert opinion: Dr. Herzke occasionally gives a dose of ceftriaxone if she is discharging a patient at 24 hours for additional coverage
- Be wary of positives after 36 hours since there is an 8-fold increase in contaminants
