A large retrospective study of how infants with fever without a cause are evaluated found that physicians are selective in deciding which of these babies will have blood, urine, or cerebral spinal fluid (CSF) cultures. Yet with or without cultures, the infants in this retrospective review had good outcomes, with no delayed identification of bacteremia or meningitis.
Researchers from Kaiser Permanente of Northern California analyzed the electronic medical records of the 1380 such infants who were born during a 3-year period and who presented for care between the ages of 7 and 90 days (14.4 infants per 1,000 full-term births). Sixty-eight percent of the febrile infants, who received care at either an emergency department (ED) or an outpatient clinic, had at least 1 culture. Older infants with lower febrile temperatures who were seen in an office setting were least likely to have a culture as were infants who did not look ill. The youngest infants, aged 7 to 28 days, were most likely to get a complete sepsis evaluation, including urine, blood, and CSF cultures. Overall, 59% of infants in this age group had a full evaluation compared with 25% of infants aged 29 to 60 days and 5% of those aged 69 to 90 days. Older infants—aged 29 to 90 days—were more likely to have a blood culture (15%) than a urine culture (12%).
Practitioners identified infections in 195 (14%) of study participants, primarily urinary tract infections (183 infants), but also bacteremia (36 infants) and meningitis (4 infants). Within 1 month of the initial evaluation, only 1% of patients returned with a urinary tract infection. No infants returned with bacteremia or meningitis.
The chart review also revealed that ill appearance was a predictor of culturing and positivity. Physicians did not obtain cultures when they believed the elevated temperature was inaccurate (for example, caused by environmental conditions) or, for older infants, when it occurred along with upper respiratory symptoms, bronchiolitis, or recent vaccination (Greenhow TL, et al. Pediatrics. 2016;138:e20160270).
Predicting which febrile infant has a serious bacterial infection is hard to do, and the younger the infant, the harder it gets. The job is made still harder in situations in which physicians have only “one shot,” as in busy EDs where families are unknown and the ability to follow up is often questionable. In those settings, where the safety net is stretched or has holes, I would still lean toward evaluation of febrile infants, especially those aged younger than 29 days. To limit the workups, check out the “Step-by-Step” protocol suggested by B Gomez and colleagues, which offers a middle ground, potentially avoiding some cultures in some infants (Pediatrics. 2016;138(2):e20154381). —Michael G Burke, MD