Understanding and managing acute encephalitis

Encephalitis is an important cause of morbidity, mortality, and permanent neurologic disability in both adults and children. The term “encephalitis” literally means inflammation of part or all of the “encephalon” or brain parenchyma. Globally, invasion by a pathogen causing direct neuronal injury is the most common cause of encephalitis. Direct injury by pathogen invasion may result in acute, subacute, or chronic manifestations. Sometimes, the inflammation is not due to invasion but is an indirect immunologically mediated injury.

Non-inflammatory diffuse dysfunction of the cerebrum is termed as “encephalopathy” and common causes include metabolic, toxic, or ischemic disorders. The term encephalopathy is also used for altered mentation due to any cause. Encephalopathy can also be acute or chronic.

Encephalitis and meningitis are overlapping syndromes. Pathologically, both viral and non-viral invasions of the brain cause some degree of both meningeal and parenchymal inflammation¹. Therefore, many clinicians prefer the term “meningoencephalitis”. Clinically, there is a spectrum of manifestations, but by and large two distinct patterns are seen². When the illness is associated with prominent sensorial alterations, the clinical syndrome is said to be “encephalitis”; when meningeal irritation is prominent, it is called “meningitis”. The term “aseptic meningitis” is used for a self-limited meningitic presentation and little or no sensorial alteration³.

The magnitude of the problem of encephalitis is difficult to assess because of a lack of uniform definitions and study populations. Worldwide the prevalence varies. In the US, there were 73 hospitalizations due to encephalitis per 100,000 population (95% confidence interval 7.1–7.6)⁴ from 2000 to 2010. In Ontario, Canada, between 2002 and 2013, the crude incidence of all-cause encephalitis was estimated to be about 4.3 cases per 100,000 persons per year⁵. The problem is further compounded by the fact that a wide variety of central nervous system disorders may mimic encephalitis. Infectious encephalitis can occur in epidemic, endemic, or sporadic form. In 2006, the World Health Organization coined the term “acute encephalitis syndrome” (AES) for the purpose of surveillance of JE. This is defined as “an illness in a person of any age and at any time of the year characterized by acute onset of fever with alteration of consciousness, stupor, coma and/or new onset of convulsions excluding simple febrile convulsions”⁶. Basically, this refers to acute illness with fever along with altered sensorium or convulsions or both. This type of presentation can have a very wide etiology. Table 1 shows the etiology of AES according to pathophysiology⁷. The prevalence and etiology of AES may vary from region to region because of geographical differences and also from time to time. In India, AES constitutes a public health problem.

In 2013, the International Encephalitis Consortium (IEC) suggested consensus criteria for case definition of encephalitis or encephalopathy of presumed infectious or autoimmune etiology⁸. Neurologic dysfunction in the form of altered mental status lasting for over 24 hours without an alternative cause must be present. Minor criteria that must be present (two for possible and at least three for probable or confirmed) include “fever ≥38°C within 72 hours, seizures, new focal neurologic findings, CSF pleocytosis (≥5 white blood cells [WBCs]/μL), neuroimaging with brain parenchymal changes or electroencephalogram (EEG) consistent with encephalitis. Definite cases require pathologic confirmation on brain biopsy, evidence of infection with a microorganism associated with encephalitis, or laboratory evidence of an autoimmune condition associated with encephalitis”⁸. Thus, the IEC case definition does not differentiate infectious from post- or non-infectious processes⁸.

Acute encephalitis syndrome etiology

The most important cause is infection—both viral and non-viral. Viral agents include primary neurotropic viruses such as arboviruses, herpesviruses, and rabies virus as well as other “incidental” nervous system pathogens such as entero, orthomyxo, paramyxo, and adenoviruses, which primarily cause disease elsewhere in the body and involve the central nervous system only occasionally⁷. Worldwide, JE is the single largest cause of viral encephalitis (VE)⁹. Non-viral brain infectious agents include bacteria, mycoplasma, rickettsiae, protozoa, metazoa, and fungi. Non-infectious brain inflammation such as in AIE, acute disseminated encephalomyelitis (ADEM), and collagen vascular disorders is another group. In infectious encephalopathies, there is a systemic infection causing altered sensorium or convulsions without invasion of the brain, thus presenting with the clinical syndrome of AES⁷. Lastly, the patient may have a functional coma (due to metabolic causes, drugs, toxins, chemicals, and so on) or a structural coma (due to tumors, space-occupying lesions, and vascular events). If there is also concurrent fever due to any other cause, such patients would present as AES⁷.

In high-income Western countries, the incidence of AIE surpasses that of any single infectious cause¹⁰. The most common sporadic infectious encephalitisis is herpes simplex virus encephalitis (HSVE)¹¹. The California Encephalitis Project enrolled 1570 patients with suspected encephalitis over a 7-year period (1998–2005): “A core battery of tests for 16 agents was performed. In addition, selective testing for other agents were performed on the basis of clinical and epidemiologic features. Only 16% of patients had a confirmed or probable etiologic agent identified, of which 69% were viral, 20% bacterial, 7% prion, 3% parasitic and 1% fungal. An additional 13% had possible etiologies identified. In this group there were many agents not hitherto implicated as a cause of encephalitis. Autoimmune etiology was found in 8% which made it more common than any single infectious agent. The remaining 63% had no etiology identified”¹⁰.

In most Asian countries, infectious causes are still far more common. In a study on VE in Chinese children, the etiology was confirmed in 52.5% of patients. The most common pathogen was human enterovirus (EV) (27.7%). The etiology of viral meningitis was identified in 42.8% of cases; the leading pathogen again was human EV (37.7%)¹². In India, JE is probably the commonest form and occurs in epidemics, especially in the south and east. Rabies is another form of VE which poses a public health problem. In the Gorakhpur division of eastern Uttar Pradesh, JE was held to be the main agent responsible for AES. There was a severe epidemic of JE in this region in 2005¹³, after which the Government of India imported the Chinese live attenuated vaccine (SA-14-14-2 strain), which was administered to children in affected districts. The proportion of AES caused by JE declined in recent years but AES itself did not decrease¹⁴. Over the last 5 to 7 years, a lot of literature has emerged from various parts of India, implicating scrub typhus (caused by Orientia tsutsugamushi) meningoencephalitis as an important cause of AES here^15–18. Nipah virus, a paramyxovirus known to cause severe encephalitis in Malaysian pig farmers, caused an outbreak of encephalitis in the southern state of Kerela^19,20. Dengue infection was known to produce neurologic manifestations, including encephalopathy, but dengue is now also recognized as a cause of VE due to actual viral invasion of the brain^21,22. In Muzaffarpur, Bihar, outbreaks of acute neurologic illness occurred which are believed to be due to toxins, namely hypoglycin and methylene cyclopropylglycine present in litchi fruit²³. A recent study from Lucknow, described trend of infectious encephalitis after immunization for JE was incorporated. Evidence of JE was found in 8.3%, dengue 7.8%, EVs 0.4%, HSV 0.8%, varicella zoster 0.4%, scrub typhus meningoencephalitis in 31.8%, Haemophilus influenza in 0.97%, and Streptococcus pmeumoniae in 0.94%²⁴.

So it is clear that a wide variety of infectious and non-infectious etiologies are associated with AES/encephalitis. However, in both Western industrialized nations and developing countries, the cause in more than half of cases remains unexplained despite extensive testing.

Epidemiology

Several factors such as age, geographical location, season, climate, and host immune status affect the epidemiology of encephalitis. Arboviruses or arthropod-borne viruses have their life cycle in insect vectors and vertebrate animals, occasionally infecting humans who are a “dead end “host²⁵. Different arboviral encephalitides have their own specific geographical distribution depending on the activity of their insect vectors. For example, arboviral encephalitides prevalent in the US include Western equine, eastern equine, Californian, and St. Louis encephalitis. Venezuelan encephalitis is found in South America, and JE in Asia. These encephalitides tend to occur in epidemics or outbreaks. Occasionally, a fresh pathogen, when introduced into a susceptible population, produces an explosive outbreak. HSVE is the commonest sporadic infectious encephalitis in Western countries. It tends to occur worldwide with little seasonal or age and sex predilection⁴. West Nile encephalitis virus was introduced in the US in the 1990s and the illness is endemic there²⁶. Mumps, measles, and rabies encephalitis have been largely eradicated from many developed countries because of effective vaccination programs.

Pathogenesis

Though poorly understood for some etiologies, a variety of mechanisms contribute to encephalitis. Encephalitis may be (i) infectious, resulting from direct invasion of the brain, most commonly gray matter by the pathogen, and (ii) immune-mediated, caused by immune-mediated damage (commonly white matter). Within the infectious group, both neurotropic and non-neurotropic (incidental) pathogens can cause encephalitis. Neurotropic viruses can cause viremia, subsequently crossing the blood–brain barrier (for example, arboviruses), or enter the brain by retrograde axonal transport (for example, rabies virus). Neuronal infection causes release of cytokines leading to cytotoxicity, inflammation, and tissue damage²⁷. Another mechanism is vasculitis leading to tissue ischemia as happens with varicella zoster virus²⁸. Michael et al. (2015) measured levels of 38 mediators in serum in 78 and CSF in 37 specimens from patients with encephalitis, 17 of whom had HSVE²⁹. The authors found that a pro-inflammatory cytokine response was associated with greater blood–brain barrier permeability, clinical severity, and damage seen on neuroimaging. Higher serum interleukin 1 receptor antagonist (IL-1RA) levels were found in patients with a good outcome (P = 0.004). These investigators therefore suggested that IL-1 antagonists should be investigated as adjunctive treatment in encephalitis²⁹.

AIE includes two major categories: (i) classic paraneoplastic LE associated with autoantibodies against intracellular neuronal antigens (for example, Hu and Ma2) and (ii) new-type AIE associated with autoantibodies to neuronal surface or synaptic antigens. Paraneoplastic LE results from an immunological response to tumor antigens, which mimic intracellular antigens in neurons. These are strongly associated with cancer, and prognosis tends to be poor because of irreversible neuronal killing by these mechanisms. The “new-type” AIE, of which anti-NMDAR induced is the commonest, occurs in association with pathogenic autoantibodies against membrane antigens or synaptic receptors. Binding of autoantibodies to their targets causes neuronal dysfunction, usually reversibly^30,31. An extracellular epitope in the GluN1 subunit of the NMDAR is recognized by these antibodies, which crosslink the NMDARs and promote internalization of the receptors. This decreases receptor density on neuronal surface and results in neuronal dysfunction. This dysfunction is reversible with removal of autoantibodies. Over a dozen new-type autoantibodies have been identified since the discovery of anti-NMDAR antibody by Dalmau et al. in 2007³². The same process is described for AMPAR encephalitis. Other autoantibodies may work through different mechanisms. For example, anti-LGI1 antibodies block binding sites of LGI1. Anti-GABAb R antibodies block the inhibitory effects of baclofen on the spontaneous firing of cultured neurons and influence receptor function³⁰.

Most of these autoantibodies are associated with specific and well characterized symptoms and the detection of these autoantibodies confirms the diagnosis. Direct brain viral infection triggering of immune-mediated disease may also coexist, as illustrated by HSV encephalitis cases with subsequent or concurrent anti-NMDAR antibodies identified³³. A recent study from New Delhi on 40 children who tested positive for autoantibodies showed that NMDAR antibodies were the commonest (60%), followed by anti-basal-gangliaantibody (17.5%), anti-GAD (7.5%), anti-Ma (5%), anti-TPO (5%), anti-Yo(2.5%), and anti-CRMP5 (2.5%)³⁴.

Clinical features

Geographic and seasonal factors, occupation, history of recent travel, contact with animals, and animal bite need to be considered. JE occurs in summer and post-monsoon epidemics in Asia. A setting of immunodeficiency may point to specific agents like cytomegalovirus³⁵.

Clinical manifestations depend on whether the brain parenchyma or meninges are predominantly involved producing an encephalitic or meningitic syndrome, respectively. However, the same agent may produce a predominantly meningitic picture in one and encephalitic picture in another patient³⁶. The severity of manifestations varies widely from mild febrile illness associated with headache to a severe disorder with convulsions, coma, neurologic deficits, and death¹. Usually, the onset is abrupt with fever and declining mental status. There may be irritability, agitation, screaming spells, confusion, delirium, drowsiness, stupor, or coma. Headache may be complained of in older children. Typical features include an initial stage of fever, headache, and vomiting lasting for less than a week followed by convulsions, coma, and neurologic deficits with or without signs of meningeal irritation. Severe cases may be associated with a life-threatening rise in intracranial tension, decerebration, or flaccid coma. Typically, this stage lasts for 7 to 10 days after which there is gradual recovery with or without sequelae⁷.

Examination should include a search for skin rash often seen with enteroviral encephalitis, measles, dengue, varicella zoster, and rickettsiae. Parotitis often occurs with mumps. Mucous membrane lesions like “cold sore”, believed to be a manifestation of herpes simplex infection, have little diagnostic value. Concurrent upper respiratory infection is characteristic of influenza. Patients with rabies may have the characteristic hydrophobia or aerophobia.

Neurologic signs in acute encephalitis do not reliably identify the underlying etiology despite the propensity of certain neurotropic viruses to affect specific focal areas of the nervous system. A constellation of frontotemporal signs with aphasia, personality change, and focal deficits or focal seizures was considered characteristic of HSE. With the availability of newer non-invasive techniques of diagnosis, milder, atypical cases and an “expanded” spectrum of HSE in children with multifocal or diffuse brain involvement have been described^37,38. Rabies may present as an ascending paralysis simulating Guillain–Barré syndrome³⁹. JE is associated with prominent extrapyramidal or basal ganglia signs in the form of rigidity and abnormal movements, especially in the convalescent stage⁴⁰.

Cerebral malaria presents as acute onset of fever with chills and rigors, bilateral upper motor neuron signs, and multisystem involvement, including acidosis, severe anemia, hepatic, renal, and pulmonary dysfunction⁴¹. ADEM is usually a monophasic illness with multifocal upper motor neuron signs along with altered sensorium.

Two distinct clinical syndromes of infectious AES are now recognized in Uttar Pradesh, India. The first is a pure neurologic illness that includes JE and HSV encephalitis. The second is a multisystem syndrome with rash, thrombocytopenia, hepatosplenomegaly, bleeding, deranged liver, and kidney function and such features are seen in dengue, rickettsial infection, enteric fever, leptospirosis, and cerebral malaria⁴².

In AIE, psychiatric manifestations are common early in the course. These may include psychosis, aggression, inappropriate sexual behaviors, panic attacks, compulsive behaviors, mutism, memory loss, euphoria, or fear. Symptoms may fluctuate rapidly. In addition, seizures, cognitive decline, coma, and abnormal movements are common^43,44. Clinical differentiation of HSE and AIE may be difficult as psychiatric manifestations and extrapyramidal movements may be seen in both. MRI, polymerase chain reaction (PCR), and antibody detection in CSF/serum may be needed.

Investigations

Diagnostic testing should include lumbar puncture, brain MRI, and EEG in all suspected cases of encephalitis. After assessment for contraindications, a lumbar puncture with opening pressure measurement should be performed to obtain CSF for total and differential cell counts, diagnostic testing, protein, and glucose. Typically, CSF in VE has essentially normal glucose, increased protein, and a mild to moderate mononuclear pleocytosis (predominantly lymphocytic but can be neutrophilic early in the course)⁴. Normal CSF WBC counts of infants are higher than those of adults. A 95th percentile cutoff of not more than 19 WBCs/μL for infants below 1 month of age or not more than 9 WBCs/μL for infants 1 to 2 months of age defines pleocytosis⁴. On the other hand, infectious encephalitis, particularly with EV (~50%) or human parechovirus (HPeV), is more likely without pleocytosis^4,8. Repeat lumbar puncture should be considered in case of persistent or worsening symptoms.

Treatment

A severe case should be managed in an intensive care unit. An active search for treatable causes of AES should continue along with measures to protect the brain from further insult. Management may be divided into supportive and specific.