a SLACK Incorporated newspaper
March 1998
NEW YORK CITY - A young infant with a fever can be a pediatrician's nightmare, admitted Keith R. Powell, MD, who has spent many years trying to find ways to help pediatricians determine which children should be hospitalized and who should not.
Powell spoke here at the 10th Annual Infectious Diseases in Children Symposium about managing these infants.
"It used to be that the basic drill for a child younger than 60 days with fever was to do a full evaluation for suspected sepsis, administer parenteral antibiotics in the hospital and wait until the cultures were negative," said Powell, who is professor of pediatrics and Noah Miller chairman of pediatrics at the Children's Hospital Medical Center of Akron, Northeastern Ohio's University College of Medicine.
"There were a number of studies to identify infants with fever who were more likely to have serious bacterial infections - so that you would not have to treat all of them. And in fact, these studies showed that children who were younger, who had high temperatures, who were ill appearing and had a focus of infection which was identifiable clinically, who had elevated white counts, an increased number of band forms or elevated sed rates were more likely to have serious infection than children who did not have these findings," he said.
However, there was a troubling finding with each study, he said. No matter which clinical and laboratory findings were grouped, each study missed at least one child with serious bacterial infection.
But full sepsis work-ups are not necessarily the norm for every young child with fever. A survey compared the differences in the management of infants with fever in a university hospital vs. a private practice. Chief residents and office pediatricians were asked how they would respond to the following child. The child was younger than 2 months old and had a temperature of 37.8° C.
Two-thirds of the residents said they would do a lumbar puncture vs. 16% of the pediatricians. In addition, half of the residents said they would hospitalize the infant vs. 9% of the pediatricians.
Powell said the difference in response may occur because of the setting. House officers see febrile infants in emergency departments. "They've never seen them before," Powell said. "They are likely never to see them again and they are not going to be followed well."
In a private office, however, the pediatrician has a relationship with the family; he or she knows the parents and whether that family will monitor the situation. In addition, the pediatrician or one of the partners is available 24-hours a day to answer a call from the parent if the child deteriorates. "I think being able to follow what happens closely is what makes the practice different," he said.
Is there a way to determine which child is at low risk? The Rochester criteria can help, Powell said. The first differential is the overall appearance of the child. A child who appears seriously ill should be hospitalized and treated. "A child who looks ill should be thoroughly managed in an urgent manner without thinking too much," Powell said.
A child who appears moderately ill or one that you are unsure of should also be hospitalized and evaluated.
![[bar]](/art/gradred.gif){kind=small}
The gray area is the young child with fever who appears generally well. If the child has skin, soft tissue or bone infections, or otitis media, the child should be evaluated, hospitalized and treated. Without those factors, do a blood count with differential, urinalysis and stool smear if the child has diarrhea.
A child with tests indicative of serious infection or who has a history of prematurity, prenatal problems, underlying conditions or has already received antimicrobial agents should be hospitalized and a complete sepsis work-up should be done.
A child without the underlying history of problems might be sent home for observation if the physician is available for consultation and the caregiver can be trusted to observe the child. The caregiver should be able to get the child to the physician or hospital within 30 minutes if conditions worsen, according to the Rochester criteria. If not, hospitalize the child.
If the child is managed as an outpatient, telephone follow-up should be done within 12 hours and a visit should be scheduled for 24 hours.
Powell told a story of a 3-week-old girl, who had a "funny spell"; her eyes rolled back into her head. The mother had described the child as "not acting well," but the description was non-specific. The pediatrician saw the baby, who appeared normal except for a temperature of 38° C. The child was sent to Powell for an evaluation.
Powell talked with the family, saw the baby and decided she could be followed. He did lab tests and gave her 50 mg/kg of ceftriaxone (Rocephin, Roche). At the 12-hour follow-up call, the baby appeared fine. Twenty-four hours later, the baby still seemed okay. Cultures were negative, and the mother did not report any more spells. "I gave a second dose of ceftriaxone and sent the baby home," he said.
Shortly after, the lab called with the finding that there were gram-negative diplococci in the blood. Powell said he called the pediatrician, gave him the information with the caveat that the baby looked great. Powell suggested that he see the baby again the next day and if she continued to do well, put the baby on five to seven days of ceftriaxone. The pediatrician insisted the baby be hospitalized, which she was and was fine.
In the end, handling the young child with fever is a judgment call.
For more information:
- Powell KR. Evaluation and management of the infant younger than 60 days of age with fever. Presented at the 10th Annual Infectious Diseases in Children Symposium. Nov. 22-23. New York.
- Jaskiewicz JA, et. al. Febrile infants at low risk for serious bacterial infection: an appraisal of the Rochester criteria and implications for management. Pediatrics. 1994;94:390-96
You can express your
views on this article, or other relevant themes, in the
Infectious Diseases in Children
Specialty Forums.