Central nervous system (CNS) infections can range in severity from self-limited aseptic meningitis syndrome to life-threatening bacterial meningitis or encephalitis. The term “aseptic meningitis” generally refers to meningitis with a lymphocytic pleocytosis for which a cause is not apparent after initial evaluation and routine stains and cultures. The majority of these patients will have resolution of symptoms without neurological sequelae, but short-term morbidity may be substantial. Meningitis due to bacteria can progress rapidly and frequently has serious long-term sequelae. Encephalitis, characterized by alterations of consciousness or other focal neurologic signs, is due to inflammation within the brain parenchyma. Since inflammation of the brain or spinal cord may also accompany meningitis, terms such as meningoencephalitis (ME) or meningoencephalomyelitis are sometimes used for these cases.

Although the annual incidence of bacterial meningitis has decreased over the last couple of decades, an estimated 4,100 cases and 500 deaths from bacterial meningitis occurred annually in the U.S. between 2003 and 2007, and the case fatality rate remains over 14%. From 1982- 1988, between 8,300 and 12,700 cases of aseptic meningitis were reported to the CDC; however, the overall incidence is probably underreported. A study in Olmstead County, MN from 1950-1981 found an adjusted incidence rate of 10.9/100,000 person-years. The majority of aseptic meningitis cases for which an etiology can be determined have ultimately been found to be due to viral infections, including enteroviruses (primarily echoviruses and coxsackieviruses), herpes simplex, West Nile virus, and varicella. In suspected viral meningitis, a specific pathogen is found in only a small minority of cases; this may in part be due to the low sensitivity of current diagnostic testing and the low frequency with which these tests are ordered in general practice.

A long list of etiologies have been identified in cases of encephalitis, including alphaviruses (Eastern, Western, and Venezuelan equine encephalitis), flaviviruses (St. Louis encephalitis, West Nile virus, Japanese Encephalitis virus), bunyaviruses (Lacrosse encephalitis, Jamestown Canyon virus), herpes simplex viruses, varicella, and Chlamydia species. Additionally, some atypical pathogens have been associated with ‘encephalopathy,’ such as Mycoplasma pneumonia. Some of these agents can cause subclinical disease or milder forms of meningitis. Many of these agents are probably underreported because they are not tested for in patients with less severe illness. Our study, which includes a wide range of severity in clinical presentations, could delineate the spectrum of illness caused by some of these agents.

Because ME is often characterized by the acute onset of severe symptoms, patients typically present to EDs for evaluation and care. If the presenting symptoms and CSF analysis are suggestive of viral meningitis, patients are frequently discharged home. Hospital-based studies of ME are therefore biased toward patients with more severe illness.

For many patients with a sudden onset of severe headache, a lumbar puncture (LP) is performed in the ED to exclude the presence of subarachnoid hemorrhage. A proportion of these patients are found to have CSF pleocytosis or other evidence of meningitis. The association of specific infectious agents with this type of clinical presentation has not beenstudied. An infectious agent could be a trigger for aneurysmal bleeding in some cases. For cases in which CSF parameters are normal, it is possible that an infectious agent may be responsible if the LP is performed early in the infectious process.

As treatment becomes available for some agents causing viral meningitis and encephalitis, identification of a specific etiology will become more important.19 Previous studies of ME have been limited by less sensitive microbiological techniques. Newer techniques, including nucleic acid-based tests, such as polymerase chain reaction (PCR), which have greater sensitivity than conventional microbiological methods, have the potential to identify an etiology in a larger proportion of cases. In addition, other newer nucleic acid testing technologies have the potential to identify novel pathogens that have not been previously described as causes of ME. For some etiologies of ME, there may not be viable pathogen or nucleic acid present in CSF, limiting the utility of testing strategies that rely on pathogen isolation or PCR of CSF. For these pathogens, serum specimens may increase the yield of identifying pathogens by either serologic testing or nucleic acid-based testing strategies.

Many previous studies of the etiology of ME have been done retrospectively. This limits the accuracy of clinical information that is available and makes it difficult to find associations between different types of clinical presentation and etiology. A prospective study can collect more accurate and detailed clinical information that could be correlated with specific etiologies.