Prompt diagnosis of bacterial meningitis is essential. The longer the delay prior to starting appropriate antimicrobial therapy, the more likely it is that the disease will cause sequelae or be fatal. Tunkel..§nd...Scheld..l.(1995) recommend that appropriate antimicrobial therapy for bacterial meningitis should be commenced within 30 min of presentation for medical care, even if delay prior to obtaining cerebrospinal fluid is unavoidable. In this situation, choice of antimicrobial is empirical and is based on the most likely causative organisms for the patient's age. Antimicrobial therapy should not be delayed whilst awaiting organization or results of imaging studies in the case of patients with focal neurological signs or papilledema. Even if lumbar puncture is delayed, the cerebrospinal fluid findings will still be helpful and a bacteriological diagnosis may be obtained by latex agglutination or polymerase chain reaction.
In patients who are immunosuppressed or postsurgical, or where cerebrospinal fluid leak is present, antimicrobial therapy should be broadened to include other Gram-negative organisms, staphylococci, or any possible suspected opportunistic organisms.
Recommendations on empirical antimicrobial therapy in bacterial meningitis have recently been adapted to take account of changing susceptibility patterns of the commonly isolated organisms.
Despite reports of penicillin-resistant meningococci in certain countries, these have not proved to be a problem in the United Kingdom or the United States. Even if they have been isolated, patients have been cured with standard doses of penicillin. Other organisms have recently developed much more worrying patterns of antimicrobial resistance. Beta-lactamase-producing Hib accounts for about 30 per cent of all isolates of this organism, and chloramphenicol-resistant Hib accounts for up to 50 per cent of isolates in some parts of the world. Fortunately, the incidence of bacterial meningitis due to Hib has dramatically decreased since the introduction of Hib vaccine.
The recent development of most concern is the growing incidence of high-level penicillin resistance in Strep. pneumoniae, with countries in Eastern Europe reporting almost 60 per cent of isolates as penicillin resistant. In the United Kingdom approximately 3 per cent of isolates are highly penicillin resistant. Alarmingly, there are now reports of high-level cephalosporin resistance, conferring resistance to the third-generation cephalosporins. Factors increasing the likelihood of infection with a resistant strain include the patient's age (under 10 or over 50 years), immunosuppression, prolonged hospital stay, infection by serotypes 14 and 23, and frequent or prophylactic use of antibiotics.
Most authorities in Europe and North America now recommend a third-generation cephalosporin such as cefotaxime or ceftriaxone as first-line treatment for bacterial meningitis. These agents are active against all common meningeal pathogens (except Listeria monocytogenes) and the common organisms are usually sensitive (except rarely Strep. pneumoniae), they penetrate excellently into cerebrospinal fluid, and they are single agents which can be given once to three times daily.
If Strep. pneumoniae meningitis is strongly suspected, there is a case for including vancomycin in empirical therapy until the antimicrobial susceptibility pattern of the isolate is determined. Vancomycin penetrates cerebrospinal fluid adequately in the presence of meningeal inflammation and its combination with a third-generation cephalosporin may be synergistic for meningitis caused by a highly penicillin-resistant organism. In addition, rifampin (rifampicin) may have a role in patients who fail to respond to this combination.
If L. monocytogenes is suspected, addition of penicillin or ampicillin is indicated. Cephalosporins have no activity against this organism. Vancomycin is active and may be adequate initial therapy whilst awaiting culture results.
In the postneurosurgical patient, initial therapy must cover Gram-negative organisms including coliforms, Pseudomonas aeruginosa, and skin flora, as well as community-acquired organisms. Broad antimicrobial coverage is essential in these patients. One possible combination is vancomycin, ceftazidime, and an aminoglycoside.
At present there are no data suggesting that quinolones or extended-spectrum macrolides are useful for empirical therapy of bacterial meningitis. However, early studies with the carbapenem meropenem look promising as single-agent therapy for a wide range of pathogens, including those causing bacterial meningitis.
The duration of antimicrobial therapy depends on the age and immune status of the patient, the etiological agent, and the clinical course or development of complications. There is no universally accepted standard. As little as 7 days of therapy or less would be appropriate for meningococcal meningitis. In Hib meningitis 10 days is the conventional wisdom, while for pneumococcal meningitis it is 14 days. Meningitis due to L. monocytogenes should be treated for 14 days, extending to 21 days in the immunocompromised host.
Duration of therapy may need to be prolonged because of complications such as development of brain abscess or subdural empyema, prolonged fever, or development of nosocomial superinfection.
Was this article helpful?