Management of bacterial meningitis in children: Controversies in the management of bacterial meningitis

There are at least 150 to 170 cases of bacterial meningitis in Canadian children younger than five years of age each year (http://cythera.ic.gc.ca/dsol/ndis/index_e.html). Streptococcus pneumoniae and Neisseria meningiditis each cause about 40% of the cases of bacterial meningitis, with other pathogens causing the remainder of cases. Haemophilus influenzae type b is a rare cause of meningitis, with just two cases identified in Canadian children in 2000, 14 years after the first vaccine was licensed (1). There are many more cases of suspected bacterial meningitis and viral meningitis that are initially treated as possible bacterial meningitis. Thus, the diagnosis of bacterial meningitis is still considered often enough that Canadian paediatricians must be aware of appropriate management considerations. In this issue of the Journal, Dr Trenna Sutcliffe (pages 449–453) reviews two important issues in the management of bacterial meningitis, namely empiric dexamethasone therapy to prevent hearing loss and routine fluid restriction to prevent or ameliorate the syndrome of inappropriate secretion of antidiuretic hormone (SIADH). She first reviews dexamethasone use and highlights the main conclusions of the meta-analysis by McIntyre et al (2), which were that a benefit from dexamethasone was more apparent overall for H influenzae meningitis than for S pneumoniae meningitis, with a benefit for S pneumoniae meningitis being apparent only when dexamethasone was given with or before the first dose of antibiotics. After consideration of the McIntyre et al (2) meta-analysis, Sutcliffe recommends routine dexamethasone use before or with the first dose of antibiotics in cases of suspected bacterial meningitis. In contrast, I would suggest that, in 2002, with H influenzae virtually eliminated, there is not sufficient evidence to support a recommendation for the routine use of dexamethasone in suspected bacterial meningitis. The McIntyre et al (2) meta-analysis did not provide conclusive results. Only S pneumoniae and H influenzae cases were considered in detail. Limited data on any degree of hearing loss that occurred in just three of 93 children with N meningiditis meningitis showed no difference between those treated with dexamethasone and those not treated with dexamethasone. There were considerable differences in the study populations and interventions that were compared. In particular, the incidence of severe bilateral sensorineural hearing loss (more than 60 dB loss or requirement for bilateral hearing aids) was highly variable in the control groups that did not receive dexamethasone (5.0% to 24.1% for H influenzae meningitis and 7.7% to 40.0% for S pneumoniae meningitis). Surprisingly, dexamethasone given with or before the first dose of antibiotics did not prevent severe hearing loss in H influenzae meningitis (odds ratio 0.53, 95% CI 0.14 to 1.94). Another issue was that the benefit of early dexamethasone therapy in S pneumoniae meningitis was based on one case of severe hearing loss in 52 children who were treated with dexamethasone, compared with nine cases in 51 control children. The statistical significance for this finding was lost when results from one study from Egypt were excluded. The Egyptian study (3) had very different characteristics, including no evaluation of hearing loss in children younger than five years of age and a higher overall S pneumoniae meningitis mortality rate (19%) than the other studies. The McIntyre et al (2) meta-analysis did not evaluate the severity of the presenting illness in the dexamethasone and control groups, and this is likely an important factor. For example, Arditi et al (4) reviewed 181 cases of S pneumoniae meningitis in children and found with univariate analysis a higher incidence of moderate or severe hearing loss in those who did not receive dexamethasone compared with those who did (46% versus 23%, P<0.05). However, when the severity of illness was controlled for in a multivariate analysis, there was no difference in hearing loss between the groups. After Dr Sutcliffe submitted her manuscript, the Canadian Paediatric Society Infectious Diseases and Immunization Committee published a statement in 2001 on therapy of suspected bacterial meningitis (5). After consideration of the published evidence and the current epidemiology of bacterial meningitis in Canadian children, the committee stated, “no recommendation for the routine use of dexamethasone for suspected bacterial meningitis can be made at this time”. The management and outcome of S pneumoniae meningitis in Canadian children from 1991 to 1999 was recently evaluated in a nested, case-control study from the Immunization Monitoring Program, ACTive (IMPACT) surveillance program comparing penicillin-susceptible cases with penicillin-nonsusceptible cases (6). One of the findings was that the use of empiric dexamethasone for suspected bacterial meningitis declined from 73% during 1991 to 1993 (when H influenzae was a more likely possibility) to just 5% from 1997 to 1999 (P<0.001). Dexamethasone therapy did not have an impact on the incidence of hearing loss or other neurological sequelae. The issue of the routine use of empiric vancomycin combined with a third generation cephalosporin for suspected bacterial meningitis is, perhaps, less controversial. Although the rate of S pneumoniae resistance to beta-lactam antibiotics is not high in most of Canada, even intermediate resistance to penicillin or third-generation cephalosporins is clinically relevant because clinical failures may occur with single drug therapy with a beta-lactam antibiotic (7). Further, although controlled clinical trial evidence is lacking, there is sufficient data from other sources to justify the routine use of combination therapy. In response to this knowledge, Canadian physicians appear to have incorporated into routine practice the use of empiric vancomycin for suspected bacterial meningitis. The case-control study referred to above (7) found that no children were given vancomycin from 1991 to 1993, but by 1999, 100% of children were given vancomycin combined with another antibiotic (6). Dr Sutcliffe presents a pertinent review of literature about the pros and cons for fluid restriction to prevent SIADH and she makes a convincing argument that children with suspected or proven bacterial meningitis should not be routinely fluid restricted, but that careful monitoring of clinical status, as well as serum sodium levels and plasma osmolality, should be performed. The example of monitoring for SIADH highlights the importance of careful monitoring of any child with meningitis to prevent or promptly detect and treat any of the myriad of acute complications that may occur, including prolonged fever, subdural effusion or empyema and seizures. The Canadian Paediatric Society Infectious Diseases and Immunization Committee's 2001 statement provides a clear approach to the drug therapy of suspected and proven bacterial meningitis (5). As well, Wubbel and McCracken (8) published a review of the overall approach to the management of bacterial meningitis in children. In addition to Dr Sutcliffe's manuscript, these publications provide helpful information for the clinician when considering the management of children with suspected or proven bacterial meningitis.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.