Study Suggests Change to the Infant Fever Work-up

No Comments
Infants with fever and positive flu don’t need the full bacteremia work-up

Several viruses are alive and well and circulating actively in your community. It seems like every other patient you’ve seen all morning has had a temperature. Wondering if you could resuscitate your retirement fund by purchasing stock in Tylenol, you turn to the resident who is ready to present the next case. Chief complaint? You guessed it. Fever.

The infant is a previously healthy, full-term 5-week old. She started with a fever last night, up to 102 at home per the mother. It is 38.7 rectally at present in the ED. She’s had some nasal congestion and infrequent cough. She’s been fussy and didn’t sleep as well last night. However, she’s still feeding avidly and there’s been no drop-off in urine output. Her past medical history is unremarkable. Her only medication has been infant Tylenol drops. This is the family’s second child, the older brother is in kindergarten and has missed several days of school for a flu-like illness. No one in the home has received a flu shot this year.


Other than the fever, the baby’s vital signs are unremarkable. The exam reveals a nontoxic and vigorous infant who looks like a peach. There is no obvious source of infection on the physical exam.

You sit down with the resident to figure out a plan. Most experts recommend a full sepsis work-up in a well-appearing, febrile five week-old without an apparent source of infection on physical examination. After all, you could miss a serious bacterial infection. But the baby could have the flu. Does that count as a source? And if it does, what does that do to the risk of bacterial infection? Do you still need to do the whole fever work-up (blood, urine, CSF)? What about a chest X-ray?

So what do we know about the risk of bacterial infections in babies with viral illness? Back in olden times, before the childhood vaccinations for HIB and pneumococcus came along, we used to get CBCs and blood cultures in kids ages 3-36 months who had fevers of > 39, looked good and had no source of infection evident on exam. We were looking for occult bacteremia (which has virtually disappeared with the new vaccines). Back then people wondered if you still had to get the blood work if the kid had a cold.  The answer was yes – they had the same rate of bacteremia as kids with no colds. In our current vaccine era, people have looked at serious bacterial infections in babies who’ve tested positive for RSV. Turns out, they don’t get bacteremic, nor do they get meningitis. Interestingly, they still can get urinary tract infections. Most of us now get urines in febrile infants with RSV, but don’t get blood or CSF.


How about flu? You’d think the same logic would apply. And it seems to. The July issue of the journal Pediatrics published a paper asking that very question. They looked at febrile babies up to 60 days of age who tested positive for flu. They saw no bacteremia or meningitis, even in the neonates (28 days and younger) who were influenza positive. However, they did see a rate of bacterial urinary infections of 2.4%. Chest X-rays were obtained at the “discretion of the attending physician in the ED”. In febrile infants that tested positive for flu the rate of lobar pneumonia on chest X-ray was 2.8% versus 8% in infants who were flu negative.

The authors end by suggesting that it is probably unnecessary to get blood and CSF in flu-positive babies, but still wise to get urine. It’s harder to say anything about chest films. Also of note, there were only a small number of babies < 28 days included in the analysis. Even though none of them had bacteremia or meningitis, the authors suggest it is still prudent to do the full sepsis work-up and admit them, regardless of the results of influenza testing. But what to do with our 5-week old peach? We got urine and a rapid flu test. We chose not to get a chest film as the baby had minimal cough, no tachypnea and clear lungs on exam. When the flu came back positive, we elected not to do blood work or a spinal tap. Because the urine looked clean we sent the infant home with close follow-up. So this flu season, you might not need to do a full sepsis evaluation in infants 29-60 days of age who test positive for influenza. You still should get urine. You can think about whether to get a chest X-ray based on symptoms and exam. But you don’t necessarily have to get blood or CSF if the baby looks good and is available for follow-up.
Krief, WI. Influenza virus and the risk of SBIs in febrile infants. Pediatrics. 2009;124(1):30-39.

Amy Levine, MD, is an associate professor of pediatric EM at UNC Chapel Hill  


Leave A Reply