| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters |
1
Department of Clinical Biochemistry, Hadassah Mount Scopus Hospital, Hebrew University, Hadassah Medical School, Jerusalem 91120, Israel
2
Department of Clinical Microbiology, Hadassah Mount Scopus Hospital, Hebrew University, Hadassah Medical School, Jerusalem 91120, Israel
aAuthor for correspondence. E-mail mayer{at}hadassah.org.il.
To the Editor:
Several reports have shown an ability of C-reactive protein (CRP) to discriminate between patients with bacterial meningitis and patients with aseptic (viral) meningitis (1)(2). Although a recent metaanalysis suggested that a negative CRP test in either cerebrospinal fluid (CSF) or serum can be used with a very high probability to rule out bacterial meningitis (3), a more recent report (4) suggested that serum concentrations are a better screening tool for this differential diagnosis.
To assess the ability of CSF CRP to differentiate gram-positive from gram-negative meningitis, we compared CRP concentrations in the blood and CSF, along with CSF nitric oxide (NO), protein, glucose, and leukocyte count, in 17 consecutive patients (age range, 2 months to 47 years) with suspected bacterial meningitis and in noninfected controls. The two patient groups had comparable CSF leukocyte counts and protein and glucose concentrations, and were not considerably different with respect to dispersion of gender and age. The noninfected controls had a similar age and gender distribution.
CRP was analyzed by a sensitive immunoturbidimetric assay using the Integra 400 analyzer (Hoffmann-La Roche). The concentration of nitrite, as an index for NO concentration, was analyzed by the Griess reaction. The CSF underwent microscopy of gram-stained smears, followed by microbiologic diagnosis. Organisms were seen microscopically in all cases, and CSF from all patients exhibited bacterial growth in culture. No patient had received antimicrobial therapy before CSF sampling. The gram-positive organisms were identified as Streptococcus pneumoniae (seven cases) and S. pyogenes (one case), and the gram-negative bacteria were Hemophilus influenzae group B (three cases), Neisseria meningitidis (five cases), and Escherichia coli (one case). Gram-negative and gram-positive cases could not be distinguished by CSF leukocyte counts (P = 0.55; Fisher’s exact test), CSF protein (P = 0.16), or CSF glucose (P = 0.11).
The mean CRP in CSF was significantly higher in patients with gram-negative bacterial meningitis as compared to patients with gram-positive bacterial meningitis (Fig. 1
). This could not be attributed to the difference in patient age between the groups because CSF CRP was not correlated with age (not shown). An increased mean CRP in the blood was also noted (3 ± 2 mg/L in controls, 163 ± 53 mg/L in gram-positive, and 272 ± 51 mg/L in gram-negative patients), but the difference between the two patient groups was statistically insignificant (P = 0.164). In the gram-negative cases, higher mean NO was also observed (8.5 ± 2.8 µmol/L vs 2.7 ± 0.9 µmol/L), but this difference had only a borderline significance (P = 0.086). Similar conclusions on significance of the CRP and NO differences were reached with the Mann–Whitney nonparametric test.
|
The ratio of CRP in CSF to CRP in blood was 3.28% ± 1.11% in gram-positive bacterial meningitis vs 9.44% ± 4.08% in gram-negative meningitis (P = 0.072; Mann–Whitney U-test).
The substantial increase in CSF CRP, as well as the trend of an increased CSF/blood ratio of CRP, suggests that infection with gram-negative bacteria enhances permeability of CRP through the blood-brain barrier. It is possible that these findings reflect the ability of the endotoxin lipopolysaccharide-s, present in gram-negative but not in gram-positive bacteria, to affect the permeability of the blood-brain barrier (5). NO may be involved in this mechanism because its concentration in CSF is higher in gram-negative meningitis. This possibility is supported by the higher potency of gram-negative bacteria to promote macrophage NO production (6), the enhanced production of NO in the CSF of septic meningitis (7), and the role of NO in permeability changes of the blood-brain barrier in LPS-induced experimental meningitis (8).
Another interesting potential explanation for the present observation is that lipopolysaccharide-s produced by gram-negative bacteria could induce local CRP synthesis in the central nervous system. CRP can be produced in neurons (9), and lipopolysaccharide-s can induce CRP in extrahepatic sites (10). This may also explain the increase, albeit nonsignificant, in serum CRP in the gram-negative cases.
There is currently no single test to diagnose the etiology of meningitis promptly and accurately. Given its high sensitivity and easy measurability, CRP may be a useful supplement for rapid diagnosis and categorization of bacterial meningitis.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. B. Chitkara, L. M. Ryan, D. Stockwell, and B. L. Wiedermann Can a Clinical Decision Rule Decrease Antibiotic Use in Viral Meningitis? Arch Pediatr Adolesc Med, December 1, 2002; 156(12): 1195 - 1198. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
