Finally, encephalitis refers to inflammation within the substance of the brain itself. This
can lead to severe parenchymal CNS damage, resulting in irreversible neurologic impairment
or death. Most often infectious, there are also rare disorders in which this occurs on a primarily
immunologic basis. Fortunately encephalitis of all types is quite rare (10,000 to 20,000 cases per
year in the United States).
Brain infections of all types are uncommon, in large part because the nervous system is
so well protected. Bacterial infection, which most typically starts as a meningitis, occurs
primarily in three settings—mechanical injury to the skull (traumatic or surgical), contiguous
untreated infection in the sinuses or mastoids, eroding through bone, or bacteremia with an
organism able to cross the blood–brain barrier.
Other organisms, particularly neurotropic viruses, have developed unique strategies to
enable CNS invasion. Herpes simplex is thought to use one of two routes—either tracking from
the olfactory epithelium to the olfactory tracts and then into the medial temporal lobes or
binding peripheral sensory nerve terminals, migrating intra-axonally to the sensory ganglia,
then tracking centrally along trigeminal branches innervating the meninges (1). Poliovirus
specifically binds receptors on motor neuron terminals, then migrates centrally within axons
(2). Other strains of organisms have developed mechanisms to cross the blood–brain barrier,
but lack the ability to bind to neurons or glia; these cause infections limited to the meninges,
and not encephalitis. In the absence of such specialized mechanisms, very few microorganisms
are capable of invading and infecting the CNS.
CLINICAL APPROACH
Given the remarkably low incidence of encephalitis in the United States, only a small subset of
febrile patients with altered mental status will actually have encephalitis. In most instances
alterations of consciousness and cognitive function will be a nonspecific response to the febrile
state, probably caused by circulating cytokines or other small molecules that cross the blood–
brain barrier and are then neuroactive (3).
Two key elements are involved in differentiating between such encephalopathies and
primary brain processes. From the systemic perspective, identification of a specific underlying
medical abnormality is the key. Neurologically, it is essential to establish whether the observed
changes are focal or not—brain disorders resulting from localized damage to the brain cause
abnormalities of function related to the site of damage. Damage to the cerebral cortex can cause
seizures, an altered level of consciousness, and cognitive difficulty. Damage to the deep white
matter causes spasticity, ataxia, visual and sensory problems, but not seizures and has a less
severe impact on alertness and cognition. Damage to the brainstem can affect level of
consciousness, long tracts that pass through the brainstem, but most importantly cranial nerve
function.
Within the brain, different functions have discrete localization as well. Damage to the
temporal lobes can cause memory and olfactory problems, frontal lobe damage affects
behavior, occipital lobe damage affects vision, etc. In brief, location dictates the specific
functions that are affected. Typically if there is a brain-damaging process, functions that are
affected remain affected throughout. In contrast, in patients with an encephalopathy
abnormalities fluctuate in space and time. Hence a detailed clinical neurologic assessment
can help differentiate between a structural process—i.e., potentially an encephalitis—and a
systemic abnormality altering the brain’s metabolic milieu, and secondarily inducing a time-
varying abnormality of brain function.
In assessing patients’ mental status, one of the first steps must be assessing language.
Without establishing meaningful communication with the patient, further assessment of brain
function can be uninterpretable. Aphasic patients are commonly described as “confused”
because what they say makes no sense. If a patient’s language sounds fluent but its content is
incomprehensible, it is understandable to interpret this as evidence of confusion. However,
several simple steps—asking the patient to follow several simple verbal commands (without
helpful gesticulations), asking him/her to name a few objects or repeat a few words—should
readily differentiate between a language disorder and a confusional state.
Similarly, the behavior of a patient with psychosis may seem inexplicable and may be
interpreted as evidence of confusion. Remarkably, although psychotic patients may
154 Halperin