Issues of Empirical Management
Timing of the Initial Dose of Antibiotic
Given the potential for neurologic morbidity and mortality, it is important to institute antibiotic therapy promptly, and the accusation of failure to treat bacterial meningitis promptly is a common reason for malpractice litigation.36 The intuitive assumption is that a delay in therapy of even a few hours affects the prognosis adversely, but the clinical data are inconclusive. Some conclusions have been inferred indirectly from observational studies comparing morbidity and mortality in patients with bacterial meningitis according to the duration of symptoms before the patient presents to the hospital. More than 20 such studies have been published; in almost half (including all 5 prospective observational cohort studies), there was no correlation between the duration of symptoms and the clinical outcome.37 Conversely, in a randomized trial comparing cefuroxime with ceftriaxone in the treatment of children with bacterial meningitis, moderate-to-profound hearing loss was more frequent (17 percent) in the cefuroxime-treated group, in which sterilization of cerebrospinal fluid was delayed, than in the ceftriaxone-treated group (4 percent).38 There are two difficulties with the interpretation of these studies. First, the remembered duration of symptoms may not accurately reflect the actual duration of meningitis. Second, the clinical outcome is affected by many variables (such as age, underlying coexisting illness, the virulence of the pathogen, and the severity of illness), and appropriate multivariable analyses to assess the independent effects of delayed therapy and sterilization of cerebrospinal fluid are lacking. Pending the appearance of data to the contrary, prompt therapy should be the standard of care.
One of the most important factors contributing to delayed diagnosis and therapy is the decision to perform cranial computed tomographic imaging before lumbar puncture.39 This practice stems from reports in the 1950s and 1960s of neurologic deterioration after lumbar puncture in patients with increased intracranial pressure or intracranial mass lesions. Proponents of the view that imaging should be done first argue that intracranial mass lesions may not be clinically evident, empirical antibiotic therapy can be instituted before imaging, and the delay in lumbar puncture does not affect diagnostic accuracy or outcome. Opponents argue that routine imaging before lumbar puncture wastes time and resources and is done instead of taking an accurate history and performing a physical examination. We believe that when acute meningitis is suspected, only patients with coma, papilledema, or focal neurologic findings require cranial imaging before lumbar puncture. If imaging is indicated, we suggest obtaining blood cultures, instituting empirical antibiotic therapy, and performing lumbar puncture immediately after the imaging if there is no intracranial mass lesion. Instituting antibiotic therapy one to two hours before lumbar puncture will not decrease the diagnostic sensitivity if the culture of cerebrospinal fluid is done in conjunction with testing of cerebrospinal fluid for bacterial antigens and with blood cultures