Cerebrospinal Fluid Cultures in Febrile Infants

Time to Positivity and Proportion Growing True Pathogens

William T. Basco, Jr, MD, MS


August 16, 2017

What's the Time to Positivity of Spinal Fluid Cultures?

Even though most infants admitted to hospitals to rule out infection don't actually have infections, the resulting hospital stays are typically 24-48 hours. Data show that 36-48 hours of observation is adequate for urine or blood cultures, but fewer data are available to show how quickly cerebrospinal fluid (CSF) cultures turn positive.

To answer this question, a recent multicenter study[1] collected data on infants ≤ 90 days of age who were seen between 2000 and 2013 in the emergency department or inpatient setting. Researchers identified infants who had a positive CSF culture, and they collected additional data, including the results of other laboratory testing and whether the infant was diagnosed with bacteremia. They also determined, from the medical record, whether the child was "ill appearing" upon presentation.

The study excluded infants whose CSF was obtained from ventriculoperitoneal shunts, patients with a history of trauma, and those undergoing any type of surgery. The study also excluded infants from whom samples were obtained more than 24 hours after admission. The researchers divided the positive CSF culture results into those that grew true pathogens and those that grew probable contaminant species.

A total of 410 positive CSF cultures were identified in infants who met the inclusion criteria. However, the organisms were considered true pathogens in only 12.9% (n = 53) of those samples, and 87.1% were judged to be contaminants. Among the true-positive CSF cultures, the median time to positivity was 25 hours (mean, 28.6 hours; 95% confidence interval, 24-33.2 hours). The median time to positivity for the contaminants was 59 hours (mean, 60.1 hours). By 36 hours, 81.1% of the true pathogen cultures were positive, and 88.7% were positive by 48 hours.

About one half (51%) of the infants with a true pathogen in the CSF were also bacteremic. Group B streptococcus comprised 51% of the true pathogens, Escherichia coli comprised 13%, and Streptococcus pneumoniae comprised 9%. Six other bacteria each comprised about 3% of the total. Of the 10 infants whose CSF grew a pathogen after 36 hours, the authors concluded that seven would nevertheless not have been eligible for early discharge (eg, at 24 hours) because of notations in the chart that they were "ill appearing." The researchers concluded that most of the true pathogens in the CSF grew positive within 36 hours but that these "true positives" represented a minority of all positive cultures.


Over the past 20 years, with increasing data from observational studies, we have definitely reduced the number of febrile infants from whom cultures are obtained, especially CSF cultures. Furthermore, with improved bacteriologic techniques, we have reduced the time we must empirically treat infants while awaiting culture results.

I have to confess, though, that these findings give me pause, because almost 19% of the true-positive CSF cultures became positive after 36 hours. It suggests that 24 hours, a "deadline" many of us have become comfortable with, may be too short. That finding certainly reinforces that outpatient follow-up with primary care providers after infants have been hospitalized to rule out an infection and a careful review of what concerning symptoms should prompt parents to return are warranted.


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.