Increase in hippocampal water diffusion and volume during experimental pneumococcal meningitis is aggravated by bacteremia

Meningitis in critically ill patients admitted to intensive care unit for severe community-acquired pneumococcal pneumonia

BACKGROUND

Although it has been reported that patients with pneumococcal pneumonia may develop meningitis, lumbar puncture is not systematically recommended in these patients, even in patients with associated bacteremia or invasive pneumococcal disease. The aim of this study was to determine the characteristics and outcomes of patients admitted to intensive care unit (ICU) for pneumococcal community-acquired pneumonia who developed meningitis.

METHODS

We retrospectively included all consecutive patients admitted to our ICU from January 2006 to December 2020 for severe pneumococcal community-acquired pneumonia according to American Thoracic Society criteria. Meningitis was defined as pleocytosis > 5 cells/mm3 or a positive culture of cerebrospinal fluid for Streptococcus pneumoniae in lumbar puncture. The primary endpoint was the proportion of patients with meningitis during their ICU stay.

RESULTS

Overall, 262 patients [64(52-75) years old] were included: 154(59%) were male, 80(30%) had chronic respiratory disease, 105(39%) were immunocompromised and 6(2%) were vaccinated against S. pneumoniae. A lumbar puncture was performed in 88(34%) patients with a delay from ICU admission to puncture lumbar of 10.5 (2.8-24.1) h and after the initiation of pneumococcal antibiotherapy in 81(92%) patients. Meningitis was diagnosed in 14 patients: 16% of patients with lumbar puncture and 5% of patients in the whole population. Patients with meningitis had more frequently human immunodeficiency virus positive status (29 vs. 5%, p = 0.02), neurological deficits on ICU admission (43 vs. 16%, p = 0.03) and pneumococcal bacteremia (71 vs. 30%, p < 0.01) than those without. The ICU mortality rate (14 vs. 13%, p = 0.73) and the mortality rate at Day-90 (21 vs. 15%, p = 0.83) did not differ between patients with and without meningitis. The proportion of patients with neurological disorders at ICU discharge was higher in patients with meningitis (64 vs. 23%, p < 0.001) than in those without. The other outcomes did not differ at ICU discharge, Day-30 and Day-90 between the two groups of patients.

CONCLUSION

Meningitis was diagnosed in 16% of patients with severe pneumococcal community-acquired pneumonia in whom a lumbar puncture was performed, was more frequent in patients with pneumococcal bacteremia and was associated with more frequent neurological disorders at ICU discharge. Further studies are needed to confirm these results.

Developing a newborn rat model of ventriculitis without concomitant bacteremia by intraventricular injection of K1 (-) Escherichia coli

BACKGROUND

Neonatal meningitis caused by Escherichia coli results in high mortality and neurological disabilities, and the concomitant systemic bacteremia confounds its mortality and brain injury. This study developed an experimental model of neonatal ventriculitis without concomitant systemic bacteremia by determining the bacterial inoculum of K1 capsule-negative E. coli by intraventricular injection in newborn rats.

METHODS

We carried out intraventricular injections 1 × 10² (low dose), 5 × 10² (medium dose), or 1 × 10³ (high dose) colony-forming units (CFU) of K1 (-) E. coli (EC5ME) in Sprague-Dawley rats at postnatal day (P) 11. Ampicillin was started at P12. Blood and cerebrospinal fluid (CSF) cultures were performed at 6 h, 1 day, and 6 days after inoculation. Brain magnetic resonance imaging (MRI) was performed at P12 and P17. Survival was monitored, and brain tissue was obtained for histological and biochemical analyses at P12 and P17.

RESULTS

Survival was inoculum dose-dependent, with the lowest survival in the high-dose group (20%) compared with the medium- (67%) or low- (73%) dose groups. CSF bacterial counts in the low- and medium-dose groups were significantly lower than that in the high-dose group at 6 h, but not at 24 h after inoculation. No bacteria were isolated from the blood throughout the experiment or from the CSF at P17. Brain MRI showed an inoculum dose-dependent increase in the extent of brain injury and inflammatory responses.

CONCLUSIONS

We developed a newborn rat model of bacterial ventriculitis without concomitant systemic bacteremia by intraventricular injection of EC5ME.