Hearing Impairment in Childhood Bacterial Meningitis Is Little Relieved by Dexamethasone or Glycerol
Peltola H, Roine I, Fernández J, et al
Pediatrics. 2010;125:e1-e8
Background and Study Summary
Hearing impairment is the most common neurologic injury sustained by children after bacterial meningitis. Although dexamethasone has shown potential in reducing the risk for hearing impairment after meningitis, the investigators contend that differing study populations in the existing trials, differing drugs used to treat meningitis, and differing methods of outcome measurement have made even meta-analyses of those trials difficult to interpret. Therefore, the use of dexamethasone as adjunctive treatment in patients with meningitis is not universally accepted. At least 1 previous trial suggested that orally administered glycerol, which is used for its hyperosmolar and diuretic effects, could reduce hearing impairment among children with meningitis.
This study was a multination, prospective, randomized, double-blind trial in children aged 2 months to 16 years with meningitis who were enrolled during 1996-2003. Children had either positive cerebrospinal fluid (CSF) culture or positive blood culture and signs and symptoms of meningitis on examination. Children with complicating conditions (such as head injury, previous neurologic impairment, or neurologic surgery) were excluded. All children received ceftriaxone as the antimicrobial agent, at a dose of 80-100 mg/kg daily for 7-10 days. Randomization was done by a factorial design, resulting in 4 groups: 95 children who received dexamethasone and glycerol, 101 children who received dexamethasone and placebo, 92 children who received placebo and glycerol, and 95 children who received placebo and placebo. Causative agents were Haemophilus influenzae type B (38%), pneumococcus (18%), and meningococcus (14%). Other bacteria accounted for 2% of isolates. Of note, 28% of the children had meningitis documented by CSF pleocytosis and chemical changes (glucose, protein) but had negative blood and CSF cultures.
Among all children, 66% experienced no hearing loss. Mild impairment occurred in 11%, moderate to severe impairment in 12%, severe impairment in 7%, and deafness in 4%. The investigators conducted analyses using 3 thresholds for hearing impairment (>40 dB, ≥ 60 dB, or ≥ 80 dB). In bivariate analyses, several factors were associated with increased risk for hearing impairment, including younger age (each additional month of age decreased hearing loss risk by 2%-6%, depending on the hearing threshold used), lower Glasgow Coma Scale at presentation (each point lower increased risk by 15%-20%), decreased CSF glucose level, anemia or granulocytopenia, and Haemophilus influenzae type B as a causative agent. In multivariate analyses, regardless of hearing threshold used, neither glycerol nor dexamethasone altered the frequency of hearing impairment or loss. Only age and Glasgow Coma Scale score were significant predictors of hearing loss, at all 3 thresholds. The investigators concluded that neither intravenous dexamethasone nor oral glycerol reduced the risk for hearing impairment among children with meningitis, regardless of the threshold used to define hearing impairment.
Viewpoint
This seems to be a very good study with many patients. The factorial design also lends strength, but I cannot tell for sure that the analysis plan took full advantage of the factorial design by combining groups for analyses, such as any dexamethasone vs no dexamethasone, an approach that would have included children receiving dexamethasone plus glycerol and dexamethasone alone in a combined group. Some of the participants received antibiotics before receiving dexamethasone, and this subset appeared to have higher risk for hearing loss at the 80-dB level. However, it is difficult to know whether this is a real finding or just a result of doing so many comparisons. In any case, as the authors point out, clinical factors had the strongest association with outcomes. It is likely that we need at least 1 other large study in which medication use and dose are standardized and in which audiologic outcomes are assessed to determine for sure that dexamethasone has limited benefit in childhood meningitis, but the investigators mention that it may be a long time before we see such additional data.
Abstract