SHM Converge

Infections in infants: An update



Converge 2021 session

Febrile Infant Update


Russell J. McCulloh, MD

Session summary

Infections in infants aged younger than 90 days have been the subject of intense study in pediatric hospital medicine for many years. With the guidance of our talented presenter Dr. Russell McCulloh of Children’s Hospital & Medical Center in Omaha, Neb., the audience explored the historical perspective and evolution of this scientific question, including successes, special situations, newer screening tests, and description of cutting-edge scoring tools and platforms.

Dr. Erin King, associate program director, University of Minnesota Pediatric Residency Program

Dr. Erin King

The challenge – Tens of thousands of infants present for care in the setting of fever each year. We know that our physical exam and history-taking skills are unlikely to be helpful in risk stratification. We have been guided by the desire to separate serious bacterial infection (SBI: bone infection, meningitis, pneumonia, urinary tract infection, bacteremia, enteritis) from invasive bacterial infection (IBI: meningitis and bacteremia). Data has shown that no test is 100% sensitive or specific, therefore we have to balance risk of disease to cost and invasiveness of tests. Important questions include whether to test and how to stratify by age, who to admit, and who to provide antibiotics.

The wins and exceptions – Fortunately, the early Boston, Philadelphia, and Rochester criteria set the stage for safely reducing testing. The current American College of Emergency Physicians guidelines for infants aged 29-90 days allows for lumbar puncture to be optional knowing that a look back using prior criteria identified no cases of meningitis in the low risk group. Similarly, in low-risk infants aged less that 29 days in nearly 4,000 cases there were just 2 infants with meningitis. Universal screening of moms for Group B Streptococcus with delivery of antibiotics in appropriate cases has dramatically decreased incidence of SBI. The Hib and pneumovax vaccines have likewise decreased incidence of SBI. Exceptions persist, including knowledge that infants with herpes simplex virus disease will not have fever in 50% of cases and that risk of HSV transmission is highest (25%-60% transmission) in mothers with primary disease. Given risk of HSV CNS disease after 1 week of age, in any high-risk infant less than 21 days, the mantra remains to test and treat.

The cutting edge – Thanks to ongoing research, we now have the PECARN and REVISE study groups to further aid decision-making. With PECARN we know that SBI in infants is extremely unlikely (negative predictive value, 99.7%) with a negative urinalysis , absolute neutrophil count less than 4,090, and procalcitonin less than 1.71. REVISE has revealed that infants with positive viral testing are unlikely to have SBI (7%-12%), particularly with influenza and RSV disease. Procalcitonin has also recently been shown to be an effective tool to rule out disease with the highest negative predictive value among available inflammatory markers. The just-published Aronson rule identifies a scoring system for IBI (using age less than 21 days/1pt; temp 38-38.4° C/2pt; >38.5° C/4pt; abnormal urinalysis/3pt; and absolute neutrophil count >5185/2pt) where any score greater than2 provides a sensitivity of 98.8% and NPV in validation studies of 99.4%. Likewise, multiplex polymerase chain reaction testing of spinal fluid has allowed for additional insight in pretreated cases and has helped us to remove antibiotic treatment from cases where parechovirus and enterovirus are positive because of the low risk for concomitant bacterial meningitis. As we await the release of revised national American Academy of Pediatrics guidelines, it is safe to say great progress has been made in the care of young febrile infants with shorter length of stay and fewer tests for all.

Key takeaways

  • Numerous screening tests, rules, and scoring tools have been created to improve identification of infants with IBI, a low-frequency, high-morbidity event. The most recent with negative predictive values of 99.7% and 99.4% are the PECARN and Aronson scoring tools.
  • Recent studies of the febrile infant population indicate that the odds of UTI or bacteremia in infants with respiratory symptoms is low, particularly for RSV and influenza.
  • Among newer tests developed, a negative procalcitonin has the highest negative predictive value.
  • Viral pathogens identified on cerebrospinal fluid molecular testing can be helpful in pretreated cases and indicative of low likelihood of bacterial meningitis allowing for observation off of antibiotics.

Dr. King is a hospitalist, associate director for medical education and associate program director for the pediatrics residency program at Children’s Minnesota in Minneapolis. She has shared some of her resident teaching, presentation skills, and peer-coaching work on a national level.

Recommended Reading

PHM groups issue Choosing Wisely® recommendations
The Hospitalist
COVID-19 in children: Weekly cases drop to 6-month low
The Hospitalist
Hospital outcomes for children with MIS-C unaffected by initial presentation site
The Hospitalist
Some things pediatric hospitalists do for no reason
The Hospitalist
The COVID-19 pandemic and changes in pediatric respiratory and nonrespiratory illnesses
The Hospitalist
Mortality trends in childhood after infant bacterial meningitis
The Hospitalist
Child suicides drive Colorado hospital to declare state of emergency
The Hospitalist
A primer on COVID-19 in hospitalized children
The Hospitalist
Improving racial and gender equity in pediatric HM programs
The Hospitalist
CDC director cites rise in hospitalizations in urging teen vaccinations
The Hospitalist
   Comments ()