EBQ:De Gans - Steroids for Bacterial Meningitis

Complete Journal Club Article

de Gans JD, et al. "Dexamethasone in Adults with Bacterial Meningitis". The New England Journal of Medicine. 2002. 347(20):1549-1556.
PubMed Full text PDF

Clinical Question

In patients with acute bacterial meningitis does dexamethasone improve outcomes when given in addition to standard antibiotic therapy?

Conclusion

Early treatment with dexamethasone in adults with acute bacterial meningitis improves outcomes, specifically neurologic sequalae in patient with Strep. Pneumonia Meningitis with no increase in GI Bleed

Major Points

• Dexamethasone may reduce the mortality of patients with S. pneumonia Meningitis. Specifically in the this study, patients with any of the following:

1. Cloudy cerebrospinal fluid
2. Bacteria in cerebrospinal fluid on Gram’s staining
3. Cerebrospinal fluid leukocyte count of more than 1000/mL

were given an empiric 10mg of Dexamethasone 20min prior or during the first dose of antibiotic administration. The goal was to both reduce mortality as well as neurologic sequelae.

• The main benefit was demonstrated with the S. pneumonia meningitis group especially those with a GCS between 8-11. However the mortality benefits have not been replicated in non developed countries and there is concern for decreased CSF penetration of antibiotics. Regardless there is a demonstrated benefit to decreased hearing loss and short term neurologic sequelae with a NNT of 21 to prevent hearing loss

*The Infectious Disease Society of America recommends the use of dexamethasone in all patients with suspected Strep. Pneumoniae meningitis. and continuation of the dexamethasone as an in-patient can be based on the culture and gram stain results of the CSF

Study Design

Prospective, randomized, double-blinded, multi-center, controlled trial

Population

• N=301; randomly assigned to 1 of 2 groups
  • 157 in dexamethasone group
  • 144 in placebo group

Patient Demographics

Group Characteristic	Dexamethasone	Placebo
Age (yr)	44+/-18	46+/-20
Bacteria + in gram stain of CSF	74%	69%
CSF WBC >1000/gram stain neg	24%	29%
Cloudy CSF only	2%	2%
Duration of symp prior to admission(median)	24	24
Seizures	10%	5%
CSF opening pressure	37+/-13	34+/-14
GCS (median)	12	12
Papilledema	7%	10%
CN palsy	9%	12%
Hemiparesis	6%	8%
CSF Strep Pneumo	37%	35%
CSF Nisseria	32%	33%
CSF other	8%	12%
CSF negative	24	24
CSF WBC (mean)	8185	7438
CSF Protein	4.3+/-3.0	4.7+/-3.2
CSF Glucose	27+/-31	27+/-29
Blood Cx +	53%	47%

Inclusion Criteria

• Age 17 yrs or older AND
• Suspected meningitis with:
  • Cloudy CSF OR
  • Bacteria in CSF on gram stain OR
  • CSF leukocyte count >1000

Exclusion Criteria

• Hypersensitivity to B-lactam antibiotics or steroids
• Pregnancy
• Cerebrospinal shunt
• Treatment with antibiotics within past 48 hours
• Active TB or fungal infection
• Recent head trauma, neurosurgery, or PUD
• Enrollment in other trial

Interventions

• Randomized to receive Dexamethasone or placebo
  • Interventions looked identical and were blinded from provider and patient
  • Given 15-20 min prior to antibiotics, but amended to allow simultaneous administration of antibiotics and intervention after interim analysis
  • 10 mg IV q6 hrs x 4 days of Dexamethasone sodium or placebo
• Initially treated with amoxicillin 2 g IV q4 x7-10 days
  • Antibiotic was changed according to susceptibility if needed

Outcomes

• Analyzed with an intention to treat analysis and last-observation-carried-forward procedure

Primary Outcome

• Glasgow Outcome Scale 8 weeks after intervention
  • Score of 5 = favorable outcome
  • Score of 1-4 = unfavorable outcome
    • Dexamethasone: 15%
    • Placebo: 25%

p value = 0.03

CI 0.37–0.94

RR 0.59

Secondary Outcomes

• Death
  • Dexamethasone: 7%
  • Placebo: 15%

p value = 0.04

CI 0.24–0.96

RR 0.48

• Focal neurological abnormalities
  • Dexamethasone: 13%
  • Placebo: 20%

p value = 0.13

CI 0.36–1.09

RR 0.62

• Hearing loss
  • Dexamethasone: 9%
  • Placebo: 12%

p value = 0.54

CI 0.38–1.58

RR 0.77

• GI bleed
  • Dexamethasone: 2 patients
  • Placebo: 5 patients

p value = 0.27

• Fungal infection

p value = 0.24

• Herpes Zoster

p value = 0.75

• Hyperglycemia

p value = 0.38

Subgroup analysis

• Causes of meningitis
  • Streptococcus pneumoniae
  • Neisseria meningitidis
  • Other bacteria
  • Culture negative CSF
• Statistically significant decrease in GOS unfavorable outcomes and death with dexamethasone in strep pneumo subgroup
• No other subgroup reached statistical significance

Criticisms & Further Discussion

• The goal of early glucocorticoids for bacterial meningitis, is to reduce the neurologic morbidity, particularly with Streptococcus pneumoniae
• Early animal research demonstrated a decrease in hearing loss in rabbits and shown in adults in Europe to reduce mortality in patient with Streptococcus pneumoniae Meningitis
• However meta-analysis demonstrated no mortality difference overall and only in the subgroup with cultures positive for Streptococcus pneumoniae and not those for Haemophilus influenzae or Neisseria meningitidis
• Dexamethasone has been shown to decrease blood-brain permeability and therefore antibiotic penetration into subarachnoid space specifically with Vancomycin However increasing the serum concentration of Vancomycin may increase the CSF levels and can possibly counteract the effects from dexamethasone
• The Infectious Disease Society of America recommends the use of dexamethasone in all patients with suspected Strep. Pneumoniae meningitis. and continuation of the dexamethasone as an in-patient can be continued based on the culture and gram stain results of the CSF
  • Suspicion for pneumococcal meningitis can be based on the inclusion criteria from the De Gans Study

Funding

Supported in part by a grant from NV Organon, which also supplied the study medication.

Sources

Authors

Colin Hoff, Daniel Ostermayer