Background. Cerebrospinal fluid (CSF) values in the noninfected neonat
e are not well delineated. Studies analyzing these values are inconsis
tent in the criteria used to define the noninfected population. The pu
rpose of our study was to examine CSF values in neonates in the first
30 days of life in whom infection was more thoroughly excluded than in
previous reports. Stringent inclusion criteria defined the noninfecte
d population, and the recently available polymerase chain reaction (PC
R) for enteroviruses was used in addition to cultures to help exclude
viral disease. Results were also stratified by age in weeks to evaluat
e for any variability that occurs in CSF values during the first month
of life. Methods. Neonates were selected from subjects enrolled in tw
o studies on aseptic meningitis. Noninfected infants were identified b
y the following criteria: (1) atraumatic lumbar puncture (less than or
equal to 1000 red blood cells/mm(3)); (2) no antibiotic therapy befor
e lumbar puncture; (3) sterile blood, CSF and urine bacterial cultures
; (4) neg ative CSF viral culture; and (5) negative CSF PCR for entero
viruses. Results. The mean +/- so total CSF white blood cell count for
108 noninfected neonates was 7.3 +/- 14/mm(3) (95% confidence interva
l 6.6 to 8.0/mm(3)) with a median of 4/mm(3) and a range of 0 to 130/m
m.(3) There were no significant differences in the mean CSF white bloo
d cell counts among age categories. Conclusions. The application of st
ringent inclusion criteria and the use of the PCR yielded a population
of infants that better represents the noninfected neonate than earlie
r reports. These values can be used for reference in evaluating the fe
brile or ill neonate.