Febrile Neonate With Probable UTI: The Likelihood of Meningitis

William T. Basco, Jr, MD, MS


August 07, 2017

Acute Bacterial Meningitis in Infants With Urinary Tract Infection

To determine the frequency of concomitant acute bacterial meningitis in neonates with febrile urinary tract infection (UTI), a recent study[1] reviewed outcomes for children evaluated from 2005 through 2013 at Texas Children's Hospital. The investigators identified positive urine cultures, then used electronic medical records to identify infants < 30 days old who presented to the emergency department with a temperature of at least 100.4°F. They collected data on all children with a UTI who also had a lumbar puncture. A UTI was defined as any culture that grew ≥ 50,000 colony-forming units of a urinary pathogen or a culture with > 10,000 colony-forming units plus a urinalysis that was highly suggestive of UTI.

Electronic health record review added demographic factors, known medical conditions (such as history of a previous UTI), circumcision status, and a measure of the clinical appearance of the child. Ill-appearing children were those whose admission documentation suggested that the child was "ill-appearing," in any sort of distress, irritable, or had a skin color change. Two definitions of acute bacterial meningitis were applied. Children with any bacterial growth on cerebrospinal fluid (CSF) culture had "definite" acute bacterial meningitis. Children with antibiotic pretreatment and elevated CSF white blood cell counts (> 20 cells/µL) but no identified viral pathogen had "probable" meningitis.

Ultimately, 236 infants who met inclusion criteria were identified for this retrospective analysis. The mean age of the infants was 18.6 days; 79% were boys; and 60% were Hispanic, 16% were non-Hispanic white, and 15% were non-Hispanic black, with other groups comprising < 5% each. Only 5.9% of the infants were considered ill-appearing, and the clinical appearance of 15.3% was undetermined. Almost 8% of the infants had been born prematurely. Among the boys, 73.3% were uncircumcised.

Among urinary pathogens, Escherichia coli comprised 86.9% of isolates, followed by Klebsiella species (4.2%) and Enterobacter species (3.4%). Among infants with a UTI, 94% had a positive urinalysis. Bacteremia was present in 9.7% of the infants. In the study cohort, no infant met the criteria for acute bacterial meningitis, and 0.8% met the criteria for probable meningitis: one with a CSF white blood cell count of 25 cells/µL and the other with 183 cells/µL. Both of those infants had been pretreated with antibiotics, and their CSF was bloody when obtained. Three additional infants had viral meningitis or meningoencephalitis.

The 95% confidence interval for having no infants (0%) with meningitis was 0%-1.6%. The authors conclude that their study suggests that acute bacterial meningitis is uncommon among infants with a febrile UTI. They note that further research is needed to identify clinical characteristics to help guide selective CSF testing in febrile newborns with UTI.


These data are important to review. They support selective CSF testing as a reasonable action in infants < 4 weeks of age who have a febrile UTI. The study authors emphasize one of the bigger limitations of their study—the fact that the diagnosis of meningitis was so rare that with only approximately 200 infants in the study, it was difficult to develop a predictive model for who might have acute bacterial meningitis.

To that caution, I would add another: Not every child seen at the hospital with a UTI underwent lumbar puncture. Testing all children for both infections would give an accurate population-based estimate of concomitant UTI and CSF infection. Instead, lumbar puncture to obtain CSF was probably selectively applied even among this population of febrile infants with UTI. This should bias toward overestimating the frequency of acute bacterial meningitis in these patients, because children who were more ill-appearing were more likely to have undergone lumbar puncture. But, we just don't know.

In the end, selective testing is reasonable, but I have to agree with the authors that we need data from a larger number of infants to refine who should have CSF testing.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.