Why did you develop the PECARN Rule for Low Risk Febrile Infants? Was there a particular clinical experience or patient encounter that inspired you to create this tool for clinicians?
For years, clinicians seeing young febrile infants have struggled with algorithms and criteria that attempt to stratify risk for invasive bacterial infections (i.e., bacteremia and/or bacterial meningitis) in young febrile infants. Clinicians do not want to miss infants with these infections, but they also don't want to test or over-treat infants who don't need it. Prior algorithms, however, are dated, use consensus-based thresholds for laboratory cutoffs, and do not use newer, more accurate biomarkers.
What pearls, pitfalls and/or tips do you have for users of the PECARN Rule for Low Risk Febrile Infants?
Although we derived the rule in infants age 0-60 days, the rule should not yet be used in those ≤28 days because of their HSV risk. This prediction rule is for invasive bacterial infections, not for serious viral infections. We also need larger validation numbers of febrile infants in the first month of life with bacteremia and bacterial meningitis before using the rule in that age group, and we strongly warned readers of that in the article. We are currently validating the score on another (approximately) 1,300 febrile infants.
Also, per the data in our supplement, we are planning to implement the cutoffs of 4,000 for ANC and 0.5 for procalcitonin, as they are easier, safer and less confusing to use clinically, and we found that they had minimal impact on the accuracy of the rule. We feel that the biggest impact of this study / prediction rule is identifying low risk infants 29-60 days old on whom we may obviate lumbar punctures, empirical antibiotic administration and hospitalization. The 28-day and younger group is a higher risk group, especially those 21 days and younger. At this point, we would advocate a full laboratory evaluation, including CSF, for those 28 days and younger. Those who are 22-28 days old and found to be low risk can perhaps have a short hospitalization or prolonged ED observation after the laboratory evaluation.
Any other research in the pipeline that you’re particularly excited about?
In addition to the large external validation of the PECARN Febrile Infant Prediction Rule, we are also currently using genomic data (RNA-based diagnostic technology) to distinguish between otherwise well appearing febrile infants with bacterial and non-bacterial infections presenting to EDs. The extensive viral respiratory panels are also expensive tests; we are therefore evaluating the impact of a positive viral panel on risk of invasive bacterial infection and whether it should impact our diagnostic evaluation in the ED and subsequent decision making.
