De Gans J, Van de Beek, D. Dexamethasone in adults with bacterial meningitis. New Engl J Med. 2002;347(20):1549-56.


What we already know about the topic: Bacterial meningitis is a dangerous condition with significant associated morbidity and mortality. Survivors often suffer neurologic sequelae. Bacterial meningitis frequently affects people who were previously healthy and high-functioning.


Why this study is important: In this randomized controlled trial, the authors examine whether administering dexamethasone along with antibiotics reduces morbidity and neurologic sequelae caused by bacterial meningitis.


Brief overview of the study: Between 1993 and 2001, the authors enrolled 301 people into the study. Eligibility criterial included age of 17 years or older, suspected diagnosis of meningitis plus cloudy CSF, bacteria on Gram staining of CSF, or CSF leukocyte count of greater than 1000 per mm3. Exclusion criteria included history of hypersensitivity to beta-lactam antibiotics or corticosteroids, pregnancy, CSF shunt, treatment with oral or parenteral antibiotics in the previous 48 hours, history of active TB or fungal infection, recent history of head trauma, neurosurgery, or peptic ulcer disease, or enrollment in another trial. Patients were randomly assigned to either receive dexamethasone (10 mg every 6 hours either just before or at the same time as antibiotics) or placebo. The primary outcome measure was the Glasgow Outcome Scale score 8 weeks after study initiation. This validated score assesses a patient’s neurologic ability/function, from 1 (death) to 5 (mild or no disability). The dexamethasone group had a 15% rate of unfavorable outcome (defined as a score less than 5 on the Glasgow Outcome Scale), while the placebo group had a rate of 25% (relative risk 0.59, p = 0.03, absolute risk reduction 10%). Subgroup analysis based on bacterial organism showed that patients with S. pneumoniae meningitis had substantial benefit from dexamethasone, while those with N. meningitides did not experience a significant benefit. Patients in the dexamethasone arm also experienced a lower rate of death than patients receiving placebo. Patients receiving dexamethasone did not have a significantly higher rate of adverse events than those in the placebo group.


Limitations: Since publication of this study, antibiotic sensitivities have changed. The study authors acknowledge that the overwhelming majority of their patients had organisms sensitive to amoxicillin and penicillin. Today, we have many more resistant bacteria. Dexamethasone decreases blood-brain barrier permeability, and may decrease penetration of vancomycin into the subarachnoid space. Also, initiation of antibiotic therapy was delayed in some cases because CT head was required before LP could be done, and doctors waited until completion of LP to initiate antibiotic therapy.


Take home points: This study suggests that dexamethasone decreases morbidity and mortality associated with bacterial meningitis without increasing adverse events like GI bleeding. If you are concerned enough about this diagnosis to perform an LP and treat with antibiotics, strongly consider ordering dexamethasone as well.