Clinical manifestations

Presentation and course

The clinical features of Japanese encephalitis virus infection range from a nonspecific flu-like illness to a severe and often fatal meningoencephalomyelitis. Children with Japanese encephalitis typically present after a few days of febrile illness, which may include coryza, diarrhea, and rigors. This is followed by headache, vomiting, and a reduced level of consciousness often heralded by a convulsion.

In a proportion of patients, recovery is rapid and spontaneous (“abortive encephalitis”). Others present with aseptic meningitis and have no encephalopathic features. The classical description of Japanese encephalitis includes a dull, flat “mask-like” face with wide unblinking eyes, tremor, generalized hypertonia, and cogwheel rigidity. These parkinsonian features were reported in 20% to 40% of Asian children and 70% to 80% of American service personnel (58; 51; 22). Rigidity spasms, particularly on stimulation, occur in about 15% of patients and are associated with a poor prognosis (51). Other extrapyramidal features include head nodding and pill rolling movements, facial grimacing, lip-smacking, opisthotonus, and choreoathetosis (51; 87).

Convulsions occur frequently in Japanese encephalitis and are reported in up to 85% of children and 10% of adults (23; 51). Multiple or prolonged seizures and status epilepticus (especially subtle motor status epilepticus) are also associated with a poor prognosis (93; 88). In a comparative study, neck rigidity, convulsions, abnormal behavior, seizures, and elevated aspartate transaminase are more frequent in pediatric patients than adults. Cerebrospinal fluid abnormalities were more frequent in adult patients (07). In a large adult study (1,282 patients) altered sensorium, convulsions, and headache were the main presenting symptoms. Movement disorders (eg, hyperkinetic movements and choreoathetoid and bizarre, ill-defined movements) were frequently encountered. The features of brainstem involvement consisted of opsoclonus, gaze palsies, and pupillary changes with waxing and waning character. Cerebellar signs were distinctly absent. Other manifestations included dystonia, decerebrate rigidity, and paralysis. Nonneurologic features of prognostic importance included abnormal breathing patterns, pulmonary edema, and upper gastrointestinal hemorrhage (79).

A subgroup of patients with Japanese encephalitis presented with a polio-like acute flaccid paralysis presentation (91). In these patients, a short febrile illness was followed by a rapid onset of flaccid paralysis in one or more limbs despite a normal level of consciousness (21). Weakness occurred more frequently in the legs than the arms and was usually asymmetrical. Thirty percent of such patients subsequently developed encephalitis, with reduced level of consciousness and upper motor neuron signs, but in the majority, acute flaccid paralysis was the only feature. At follow-up there was persistent weakness and marked wasting in the affected limbs. Nerve conduction studies demonstrated markedly reduced motor amplitudes, and electromyography showed chronic partial denervation, suggesting anterior horn cell damage (91). Flaccid paralysis also occurs in patients with “classical” Japanese encephalitis. It is reported in 5% to 20% of comatose patients (23) and is associated with abnormal signal intensity on T2 weighted MRI images of the spinal cord (54).

Wang and colleagues reported data on 47 patients with Japanese encephalitis who presented with Guillain–Barré syndrome (101). CSF evaluation, in 38 patients, revealed albumin-cytologic dissociation. In the majority (47%) of patients, Guillain-Barré syndrome was of acute motor axonal neuropathy variety. In 18 cases Guillain-Barré syndrome was of acute motor-sensory axonal neuropathy type. Acute inflammatory demyelinating polyneuropathy and acute sensory neuropathy verities were noted in four and three patients, respectively. Twenty-eight patients were treated with intravenous immune globulin. Half of these responded well to intravenous immune globulin treatment (101).

A report from Indonesia revealed that Japanese encephalitis virus infection was even prevalent in patients with non-encephalitic acute febrile illness (63). Serum specimens of 144 patients with acute febrile illness was tested for anti-Japanese encephalitis virus IgM antibody. Twenty-six (18%) patients either had confirmed anti-Japanese encephalitis virus IgM antibody positivity or demonstrated equivocal results. Five (3.5%) and eight (5.6%) patients, respectively, were able to fulfill the criteria for confirmed or probable Japanese encephalitis infection (63).

Prognosis and complications

The case-fatality rate among those with symptomatic Japanese encephalitis can be as high as 30%. Permanent neurologic or psychiatric sequelae can occur in 30% to 50% of patients (10; 106). Fixed flexion deformities of the arms, and hyperextension of the legs with “equine feet” are common. Twenty percent of patients have severe cognitive and language impairment (most with motor impairment too) and 20% have further convulsions (40). A higher rate of sequelae is reported for children than adults (82). More detailed studies have suggested that approximately half of those who were classed in the “good recovery” group have more subtle sequelae such as learning difficulties, behavioral problems, and subtle neurologic signs (52). The long-term prognosis of Japanese encephalitis was assessed in Malaysia. In this study 118 patients were recruited. Ten patients (8%) died during the acute phase of illness. At hospital discharge, 44 (41%) of the 108 patients who survived had apparent full recovery; 33 (31%) patients had severe neurologic sequelae (73). On long-term follow-up, more than one half of the patients continued to experience neuropsychological sequelae and behavioral disorders. A combination of poor peripheral perfusion, with prolonged capillary refill time (greater than 2 sec), Glasgow coma score (8 or less), and two or more witnessed seizures predicted a poor long-term outcome with 65% sensitivity and 92% specificity (73).

Chen and colleagues evaluated long-term neurologic complications and mortality in 352 adult patients with Japanese encephalitis (15). Stroke was the most common neurologic complication, followed by epilepsy and convulsions, encephalopathy and delirium, and parkinsonism. A study was conducted on 328 children with Japanese encephalitis to determine the risk factors of postencephalitic epilepsy and construct a risk prediction model. Sixty five point nine percent of the children had seizures in the acute phase, and 14.7% developed postencephalitic epilepsy. Seizure number greater than 5, status epilepticus, and coma were identified as independent risk factors for postencephalitic epilepsy (14).

Japanese encephalitis is a risk for travelers who visit Japanese encephalitis-endemic countries. Hills and colleagues reviewed all published Japanese encephalitis cases in travelers from nonendemic areas from 1973 through 2008 and assessed factors related to risk of infection. The authors collected 55 cases that occurred in citizens of 17 countries. Ten (18%) persons died, and 24 (44%) had mild to severe sequelae (35).

Clinical vignette

A 13-year-old Vietnamese boy presented with progressive leg weakness. On the first two days of illness, he had fever, headache, vomiting, and neck stiffness. By day three he had progressive weakness of the left leg, which was painful. This was followed by weakness and pain in the right leg and lumbar spine. By day five he was unable to walk or sit unsupported and was brought to a hospital. He had no past symptoms or family history of note, and had been fully vaccinated against diphtheria, tetanus, and polio. On examination he was fully conscious, with neck stiffness. His temperature was 40.3 C; pulse was 130 beats per minute, and respiratory rate was 38 breaths per minute. The cardiovascular, respiratory, and abdominal examination was otherwise unremarkable. The cranial nerves and upper limbs were completely normal. He had marked flaccid weakness of both legs with the Medical Research Council grade 0/5 in all groups except the ankles and great toes, which were 1/5. Deep tendon reflexes in the legs were absent, as were the abdominal and cremasteric reflexes. Light touch, pinprick, and joint position sensation were normal in the legs, but the skin and muscles were tender. Perineal sensation was normal.

Thick and thin blood films were negative for malaria parasites. Hematocrit was 37%; there was a peripheral leukocytosis with 16,000 white cells/mm3 (69% neutrophils). Platelets, urea, and electrolytes were normal. A lumbar puncture had an opening pressure of 100mm CSF, two white cells/mm3, and 84 red cells/mm3; protein was elevated at 74 mg/dL, and the glucose was 51mg/dL (plasma glucose 51mg/dL). Gram stain and bacterial culture were negative. A chest x-ray was normal, Mantoux test was negative, and stool culture was negative for enteroviruses, including poliovirus. An IgM capture dot enzyme immunoassay for Japanese encephalitis virus was positive in both CSF and serum (92); this was confirmed by an IgM ELISA, which showed 186 units in the CSF and 153 units in the serum (normal levels are less than 40 units).

His fever continued, and on day seven of the illness he became confused (Glasgow coma score eyes=3, motor=5, verbal=4). There were no clinical signs of raised intracranial pressure and no new focal signs. A guide airway was inserted, and he was fed via a nasogastric tube. By day 10 he was fully conscious, and motor function was slowly improving. When he was discharged from the hospital 2 months later, he could walk with a waddling gait. At a 1-year follow-up, his gait was still abnormal and there was muscle wasting, more marked in the right leg than the left (90).

Biological basis

Etiology and pathogenesis

Japanese encephalitis is caused by the flavivirus Japanese encephalitis virus, which has a small (50 nm) lipoprotein envelope surrounding a nucleocapsid comprising of core protein and 11 KB single stranded RNA (3800 kD). There are at least five genotypes of Japanese encephalitis virus in Asia (17; 16; 98) and several closely related neurotropic flaviviruses occur across the globe (87; 85). In Australia, Murray Valley encephalitis virus causes sporadic encephalitis cases and occasional epidemics, and St. Louis encephalitis virus causes a similar disease in America. West Nile virus was previously confined to Africa, Europe, and the Middle East, but arrived in North America in 1999 (08) and has since spread across the continent to cause large outbreaks (74).

The nucleotide sequence of Japanese encephalitis virus includes 5’ and 3’ untranslated regions, genes for seven non-structural proteins and three structural proteins. One of these, the envelope protein, is the major component of the surface projections of the virion and is thought to bind to cells and mediate membrane fusion and cell entry (71). A highly sulphated heparan sulphate molecule has been identified as the putative receptor of flavivirus cell entry (18). In animal models of Japanese encephalitis, the envelope protein plays a role in determination of virulence phenotypes, but whether this is important in determining the clinical presentation of human infections is unknown.

Almost all children living in endemic areas become infected with the Japanese encephalitis virus, but only about 1 in 25 to 1 in 1000 develops clinical features (99). Factors determining which of those infected will develop disease are unknown but could include viral factors such as route of entry, titer, and virulence or host factors such as age, genetic make-up, general health, and preexisting immunity. An observation noted that TLR3 gene polymorphism might confer host genetic susceptibility to Japanese encephalitis in Indian population (06).

In endemic areas, humans become infected with Japanese encephalitis virus after mosquito bites. The Japanese encephalitis virus multiplies locally and in regional nodes. After a phase of transient viremia, invasion of the central nervous system occurs via hematogenous spread. In the neurons, the virus replicates and matures in the neuronal secretory system, mainly the rough endoplasmic reticulum and Golgi apparatus, eventually destroying these structures (97). In an experimental study, authors demonstrated that the permeability of the blood-brain barrier changed in response to viral infection, leading to the entry of Japanese encephalitis virions or putatively infected leukocytes from the periphery to the cerebrum as the initial site of infection in the central nervous system (61).

At autopsy, CNS findings in Japanese encephalitis reflect the inflammatory response to widespread neuronal infection with the virus (69; 45). The thalamus, basal ganglia, midbrain, cerebellum, and anterior horns of the spinal cord are heavily affected, providing anatomical correlates for the tremor, dystonias, and flaccid paralysis that characterize the disease. Pathologic changes in the brains of patients with acute Japanese encephalitis are characterized by glial nodules and circumscribed necrolytic foci. Focal regions of white matter involvement are also present (30).

Invasion of neurons by the Japanese encephalitis virus is followed by perivascular cuffing, infiltration of inflammatory cells into the parenchyma, and phagocytosis of infected cells. There may be no histological signs of inflammation in patients who die rapidly, but immunohistochemical studies reveal viral antigen in morphologically normal neurons (45). This may explain the normal CSF findings in a proportion of patients with Japanese encephalitis. One study used routine histological staining, immunohistology, and electron microscopy to examine brain material from four fatal human cases and made comparisons with material from a mouse model (27). In human material there was edema, perivascular inflammation, hemorrhage, microglial nodules, and acellular necrotic foci, as has been described previously. In addition, new evidence was suggestive of viral replication in the vascular endothelium, with endothelial cell damage. Viral antigen was also found in neurons.

Both humoral and cellular immunity appear to be important in Japanese encephalitis. The humoral immune response has been well characterized. In primary infection (ie, when Japanese encephalitis virus is the first flavivirus with which an individual has been infected) a rapid and potent IgM response occurs in serum and CSF within days of infection. Most patients have elevated titers by day 7 and attempts to isolate virus are usually negative in such patients (09). However, the failure to mount an IgM response is associated with positive virus isolation and a fatal outcome (55; 57). Asymptomatic infection with Japanese encephalitis virus is also associated with elevated IgM in the serum, but not CSF (09). In patients with secondary infection (ie, those who have previously been infected with a different flavivirus such as dengue infection or yellow fever vaccination), there is an anamnestic response to flavivirus group common antigens (09). This secondary pattern of antibody activation is characterized by an early rise in IgG with a subsequent slow rise in IgM.

The cellular immune response appears to contribute to the prevention of disease in animal models of Japanese encephalitis by restricting virus replication before the CNS is invaded. In humans, specific T cell responses (including CD8+ T lymphocyte proliferation and CD4+ T lymphocyte responses that recognize the envelope protein in a HLA restricted manner) have been demonstrated in convalescent patients with Japanese encephalitis and vaccine recipients (01). Proliferate responses to the virus’ PrM, E, NS1, NS3 and NS5 have been reported for T cells from humans that had previously had subclinical infections with Japanese encephalitis virus (49). Memory T cells responses to NS-3 were especially common, with IFN-g production in both CD4+ and CD8+ T cells consistent with a Th1-type response (48). Interestingly, patients six months after hospital discharge had similar proliferative responses to healthy Japanese encephalitis virus-exposed individuals but produced less IFN-g (measured by ELISA) (47).

Observations suggest that the increased microglial activation following Japanese encephalitis virus infection influences the outcome. It is likely that the increased microglial activation triggers bystander neuronal damage (28). Experimental studies suggest that neuronal apoptotic death and activation of microglial cells and astrocytes play a crucial role in the pathogenesis of Japanese encephalitis. The Japanese encephalitis virus induces neuronal apoptotic death and release of cytokines that initiate microglial activation and release of proinflammatory and apoptotic mediators with subsequent apoptotic death of both infected and uninfected neurons. Activation of astrocytes and microglial and endothelial cells likely contributes to inflammatory cell recruitment and blood-brain barrier breakdown (72). Proinflammatory cytokine and chemokine levels in the CSF are associated with a bad outcome in patients with Japanese encephalitis (102).

Epidemiology

Culex mosquitoes transmit Japanese encephalitis virus in an enzootic cycle that involves pigs, chickens, and other birds. The most important vector for human infections is Culex tritaeniorhynchus, which breeds in pools of stagnant water such as rice paddy fields (42). Humans become infected with Japanese encephalitis virus coincidentally when living or traveling in close proximity to the virus’ enzootic cycle. In rural Asia, the majority of the population is infected during childhood or early adulthood (37). Approximately 10% of the susceptible population is infected each year but most infections of humans are asymptomatic or result in a non-specific flu-like illness. Humans are considered the dead-end host, as the brief periods of viremia and low viral titers do not facilitate further transmission of the virus.

Japanese encephalitis is the most important cause of viral encephalitis in many countries of Asia, with nearly 68,000 clinical cases every year. The annual incidence of clinical disease varies (both across and within countries), ranging from less than 10 to more than 100 per 100,000 population. The case-fatality rate can be as high as 30%. Permanent neurologic or psychiatric sequelae can occur in 30% to 50% of patients. Twenty-four countries in the World Health Organization South-East Asia and Western Pacific regions have endemic Japanese encephalitis transmission, exposing more than 3 billion people to risks of infection. Japanese encephalitis primarily affects children. Most adults in endemic countries have natural immunity after childhood infection, but individuals of any age may be affected (10; 106). In northern areas (northern Vietnam, northern Thailand, Korea, Japan, Taiwan, China, Nepal, and northern India) huge epidemics occur during the summer months, whereas in southern areas (southern Vietnam, southern Thailand, Indonesia, Malaysia, Philippines, Sri Lanka, and southern India) Japanese encephalitis tends to be endemic: cases occur sporadically throughout the year with a peak after the start of the rainy season (99). These differences probably relate to climate and vector abundance, but viral genotypes and host susceptibilities may also be important. Japanese encephalitis surveillance has been conducted since 1965 as a part of the National Epidemiological Surveillance of Vaccine Preventable Diseases in Japan. Over 1,000 Japanese encephalitis cases were reported annually in the late 1960s. The number of Japanese encephalitis cases has since markedly decreased, with less than 10 cases reported annually from 1992 to 2004. A total of 361 Japanese encephalitis cases were reported between 1982 and 2004. Prognosis was available for 320 cases; 58 (18%) died, 160 (50%) recovered with neuropsychiatric sequelae, and 102 (32%) completely recovered. Seventy-eight percent of these cases were 40 years old or over with a peak age group of 60 to 69 years of age. Japanese encephalitis predominantly occurred in unvaccinated populations (03). One report described three cases of Japanese encephalitis among United States travelers returning from Asia (11). All were Asian immigrants or family members who traveled to Asia and had not been vaccinated for Japanese encephalitis. These three patients experienced fever with mental status changes, but Japanese encephalitis was recognized early in the clinical course of only one patient. All recovered, but two patients had residual neurologic deficits. Data from an endemic region noted that 26.9%, 9·9%, and 14.8% of acute encephalitic syndrome cases were positive CSF serology for Japanese encephalitis in the years 2011, 2012, and 2013, respectively. Of the total Japanese encephalitis confirmed cases, 30% were adults. Males were more commonly affected than females. Japanese encephalitis outbreak was seen in the monsoon and post-monsoon season (43). In Uttar Pradesh, a heavily populated province of North India, there were 47,509 reported cases of acute encephalitis syndrome between 2005 to 2018 (83). Among patients of acute encephalitis syndrome, approximately 10% had confirmed Japanese encephalitis.

Japanese encephalitis is, generally, considered a disease of young children (107). A Chinese report noted that even older adults are equally liable to be affected with Japanese encephalitis. In 2018, 1800 Japanese encephalitis cases were reported in China; 64% of these cases were adults aged 40 years or older.

The geographical area affected by Japanese encephalitis virus has expanded in the last 50 years. Epidemics have occurred in China since 1935, in South Korea since the late 1940s, and in Vietnam and Thailand since the mid-1960s (99). The disease was recognized in southern India in 1955, but since the 1970s large outbreaks have occurred further north. Epidemics in Nepal have confirmed the continuing spread of Japanese encephalitis in a northwestern direction (02). The disease has also spread across Southeast Asia and the Pacific Rim, reaching mainland Australia for the first time in 1998 (32).

The reasons for the spread of Japanese encephalitis are not completely understood, but may include changing agricultural practices, such as increasing irrigation (which allows mosquito breeding) and animal husbandry (which provides host animals). In developed countries such as Japan, Taiwan, and South Korea, the number of cases of Japanese encephalitis has fallen, probably due to a combination of factors including mass vaccination of children and socioeconomic changes (87). However, in Korea the widespread use of vaccine in children has been associated with a higher incidence of Japanese encephalitis in those over 15 years of age (99).

An outbreak of Japanese encephalitis virus has occurred in Australia, with 40 confirmed or probable human cases reported, including five deaths. This is the first Japanese encephalitis outbreak in Australia since 1998, and it has been linked to genotype 4 Japanese encephalitis closely related to strains in Indonesia. The introduction of Japanese encephalitis into Australia is thought to be facilitated by infected migratory birds, mosquitoes, or both, with transmission to pigs leading to virus amplification. Contributing factors to the outbreak include La Niña weather conditions, intensive pig farming, and a large feral pig population. Mosquitoes, particularly Culex annulirostris, are the primary vector, presenting challenges for identifying at-risk groups and controlling the spread of the virus (100; 108).

A paradigm shift in epidemiology has been observed as Japanese encephalitis is becoming increasingly more common among adults due to expanding Japanese encephalitis transmission areas, improved Japanese encephalitis control in children leading to more adult cases, and increased tourism to Japanese encephalitis risk areas. Clinical presentation and severity of illness are similar for both adults and children, with the only clear difference being a lower frequency of seizures in adults. Childhood vaccination programs in endemic areas should be strengthened and maintained to reduce the burden of disease in both children and adults (36).

Prevention

Measures to control Japanese encephalitis include those that interfere with the virus’ enzootic cycle and those that prevent human infection, including vaccines. Attempts to control breeding of Culex mosquitoes and vaccination of pigs have mostly proved ineffectual. Efforts to reduce the number of Culex bites, such as minimizing exposed skin and using insect repellents containing DEET (N,N-diethyl-3methlybenzamide) are advisable for short-term visitors, but are often not practical for permanent residents. All travelers to Japanese encephalitis-endemic areas should take precautions to avoid mosquito bites to reduce the risk for Japanese encephalitis. Personal preventive measures, like use of repellents, long-sleeved clothes, and coils and vaporizers, are recommended (106).

The most widely used vaccine for Japanese encephalitis virus is a mouse brain-derived vaccine developed in Japan and based on the Nakayama or Beijing-1 strain. Its efficacy was shown in large double-blind randomized tetanus-toxoid controlled trials in Taiwan and Thailand involving more than 300,000 children (37; 81). Although it is associated with a moderate frequency of mild side effects, serious neurologic adverse events are very rare. Encephalitis, encephalopathy, seizures, or peripheral neuropathy occur in approximately one per million recipients. However, since 1989 a new pattern of allergic reactions has been described (itching, urticaria, and rarely angioedema of the face), which is more likely in those with a previous history of allergy and occasionally may require hospitalization and corticosteroid therapy (75). The inactivated mouse brain-derived vaccine is now commonly replaced by cell culture-based vaccines (105) (see Table 1).

Since the late 1980s, a live attenuated Japanese encephalitis vaccine (produced by passing the virus through weanling mice, then culturing in primary baby hamster kidney cells), has been given to over 100 million children in China. It is safe and immunogenic. In a case control study, two doses one year apart were shown to be 97.5% effective (34); in another study, a single dose of vaccine was shown to be an efficacious prevention technique when administered only days or weeks before exposure to infection (05). The vaccine was licensed in China in 1988 and is being used in other Asian countries, but it is not yet available in the West. In addition, a chimeric vaccine is being developed in which the structural proteins of the long-established 17D yellow fever vaccine are replaced with the same proteins from the live attenuated Japanese encephalitis vaccine (04). It has been shown to be safe and immunogenic (70). In a Cochrane review, authors have concluded that because only one of three vaccines have been directly investigated for effectiveness in a randomized controlled trial, it is not possible to compare the effectiveness of currently used vaccines in the prevention of clinical disease (81).

Table 1. Vaccines for Japanese Encephalitis

In a multicentric randomized-controlled study, Yan Li and colleagues demonstrated that simultaneous administration of live-attenuated Japanese encephalitis vaccine and measles and rubella vaccine does not adversely affect measles and rubella seropositivity. Therefore, both vaccines can be administered together in countries where Japanese encephalitis is endemic (56).

Japanese encephalitis vaccine is recommended for international travelers who intend to spend a month or longer in endemic areas during the Japanese encephalitis virus transmission season and for laboratory workers with a potential for exposure to infectious Japanese encephalitis virus (25). However, considering the variability of Japanese encephalitis from year to year, its unpredictability, and the poor reliability of some epidemiological data, identifying areas of epidemic transmission is difficult, and wider use of the vaccine has been advocated (42; 11). A newer purified, inactivated Japanese encephalitis virus vaccine (IC51) has been developed, which is manufactured in a Vero cell culture substrate. Studies show that the vaccine is both safe and immunogenic. This vaccine is currently aimed for international travelers (44).

Adverse events following receipt of inactivated mouse-brain-derived Japanese encephalitis vaccine reported to the United States Vaccine Adverse Event Reporting System from 1999 to 2009 were reviewed (59). During this period, 300 adverse event reports following vaccination were received (24 per 100,000 doses distributed). Out of this, 106 (35%) were classified as hypersensitivity reactions (8.4 per 100,000 doses), and four (1%) were classified as neurologic events (0.3 per 100,000 doses). Only 23 reports (8%) described serious adverse events (1.8 per 100,000 doses distributed).

A study from China observed that high incidence of childhood onset myasthenia gravis in Chinese population is linked to universal live-attenuated Japanese encephalitis vaccination. The authors of this study suggested that live-attenuated Japanese encephalitis vaccine induced an autoimmunity against the acetylcholine receptor through mechanism of molecular mimicry (33).

A One Health approach that integrates human, animal, mosquito, and environmental surveillance is crucial to understanding the epidemiology of Japanese encephalitis virus. Although human case surveillance provides limited early signals of transmission, mosquito and animal surveillance can provide early warning signals that enable the implementation of public health measures. Travelers to endemic areas should be made aware of the risk and advised on vaccination and mosquito bite avoidance measures. The evolving epidemiology of Japanese encephalitis virus in Australia has implications for the possible threat the disease may pose to other regions (64).

Differential diagnosis

Confusing conditions

Infection with Japanese encephalitis virus may be asymptomatic or may cause febrile illness, meningitis, myelitis, or encephalitis. Encephalitis is the most commonly recognized presentation and is clinically indistinguishable from other causes of an acute encephalitis syndrome. Clinically, a case of acute encephalitis syndrome is defined as a person of any age, at any time of year, with the acute onset of fever and a change in mental status (including symptoms such as confusion, disorientation, coma, or inability to talk) and new onset of seizures (excluding simple febrile seizures). Other early clinical findings may include an increase in irritability, somnolence, or abnormal behavior greater than that seen with usual febrile illness (104). Lymphocytic CSF with an elevated protein and a normal glucose ratio supports the diagnosis, but other causes must be excluded, and virological confirmation is needed. The differential diagnosis is broad, particularly in adults, and can be divided into viral encephalitides, nonviral CNS infections, infectious diseases with CNS manifestations, and noninfectious diseases:

Japanese encephalitis may be confused with other arboviral encephalitides, particularly in areas where more than one virus is endemic. The flaviviruses, Murray valley, West Nile, and even Dengue can cause similar illnesses (41; 89). Outbreaks of Nipah virus encephalitis (20), human enterovirus 71 (103), and Chandipura virus (77) have shown the need to be cautious before attributing outbreaks of encephalitis to Japanese encephalitis. Distinguishing Japanese encephalitis from partially pre-treated bacterial meningitis and cerebral malaria may be particularly difficult, because in many parts of the tropics antibiotic and antimalarial drugs are readily available without prescription. On the basis of their observations in an endemic region, Dubot-Pérès and colleagues cautioned that detection of anti-Japanese encephalitis virus IgM in patients with acute encephalitic syndrome may be deceptive, and it is advisable to consider other treatable diseases, particularly bacterial infections before labelling the diagnosis of Japanese encephalitis (24). In 6% of CSF anti-Japanese encephalitis virus IgM positive cases, evidence of a pathogen other than Japanese encephalitis was confirmed.

Garg and co-workers described an unusual case of acute encephalitis syndrome positive for immunoglobulin-M antibodies (in serum as well as in cerebrospinal fluid) against both dengue virus and Japanese encephalitis virus. Co-positivity in an endemic region (both for dengue and Japanese encephalitis) is possible either because of co-invasion or because of co-detection resulting from well-described cross reactivity of serological tests between dengue virus and Japanese encephalitis virus (26). In one case, the patient had MRI evidence of bilateral thalamic and brainstem involvement, and serology was positive for both dengue and Japanese encephalitis (84).

Acute flaccid paralysis due to Japanese encephalitis virus must be distinguished from other viruses that attack anterior horn cells (poliovirus, nonpolio-enteroviruses, and West Nile virus) and from post-infectious causes of weakness such as Guillain-Barré syndrome. Poliomyelitis and Japanese encephalitis virus myelitis may be clinically indistinguishable, although the presence of bladder involvement suggests the latter (91). Of the enteroviruses that cause flaccid paralysis, enterovirus 71 is the most important. It is especially likely if paralysis occurs during an outbreak of hand foot and mouth disease (66; 39). Differentiating Japanese encephalitis virus myelitis from the acute inflammatory demyelinating polyneuropathy form of Guillain-Barré syndrome is not normally difficult, because the latter typically occurs sometime after an acute febrile illness and has symmetrical sensory and motor nerve involvement. Distinguishing the acute motor axonal neuropathy form of Guillain-Barré syndrome is more difficult because it too affects only lower motor nerves, and large epidemics have been described in Asia (65). However, it is usually symmetrical and post-infectious rather than occurring during an acute febrile illness.

Japanese encephalitis infection, like herpes simplex encephalitis, can trigger anti-N-methyl-d-aspartate receptor (NMDAR) immunoglobulin G (IgG) synthesis and autoimmune encephalitis (19; 96). A study by Ma and colleagues revealed that among 63 patients with Japanese encephalitis, five patients relapsed during the convalescence phase (62). Four younger patients presented with choreoathetosis, and one patient presented with the psychiatric and behavioral symptoms. NMDAR antibodies in the CSF of three patients were positive, and they were diagnosed with anti-NMDAR encephalitis. In all other patients, anti-N-methyl-d-aspartate receptor antibodies testing was negative. Liu and colleagues reported a series of 31 children with Japanese virus encephalitis who were prospectively studied for development of antineuronal autoantibodies during the course of their encephalitis (60). Of 31 children, five developed an apparent relapse of their encephalitis: of these five children, two had developed anti-NMDAR antibodies, one anti-γ-aminobutyric acid-B receptor (GABABR) antibodies, and two antibodies to unknown neural membrane antigens. The exact pathogenesis of the virus encephalitis induced N-methyl-D-aspartate receptor or other membrane encephalitis is not exactly known. It has been hypothesized that synaptic autoimmunity develops following release of antigens because of virus induced neuronal damage. Coinfection of neurocysticercosis and Japanese encephalitis has frequently been noted in the endemic regions, often resulting in poor outcome (31; 95).

Diagnostic workup

In most patients with Japanese encephalitis, there is a peripheral neutrophil leukocytosis. Typically, there is a moderate CSF pleocytosis of 10 to 100 cells per mm3, with predominant lymphocytes, mildly elevated protein (50 to 200 mg%) and a normal glucose ratio. However, polymorphonuclear cells may predominate early in the disease, or there may be no CSF pleocytosis.

Computed tomography may show bilateral nonenhancing low-density areas in the thalamus, basal ganglia, midbrain, pons, and medulla (68). The most consistent MRI findings in Japanese encephalitis are bilateral thalamic lesions with or without hemorrhagic changes. Lesions are also noted in the substantia nigra, brainstem, cerebellum, cerebral cortex, and white matter. Lesions in the substantia nigra may be characteristic of patients with parkinsonian features (76). On MRI, there may be more extensive changes and characteristic thalamic lesions of mixed intensity on T1 and T2; weighted scans are suggestive of hemorrhage (54). These changes may be useful in distinguishing Japanese encephalitis from Herpes simplex encephalitis, where the changes are characteristically frontotemporal. Medial temporal MR lesions have been reported in several other diseases of central nervous system-like paraneoplastic limbic encephalitis and neurosyphilis.

In Japanese encephalitis, medial temporal lesions are often associated with bilateral thalamic or midbrain involvement. Temporal lobe involvement in Japanese encephalitis dominantly affects the hippocampus (30; 46).

A variety of electroencephalographic abnormalities have been reported in Japanese encephalitis including theta and delta coma, burst suppression, epileptiform activity, and occasionally alpha coma, but none seem to be diagnostic.

Specific tests for diagnosing Japanese encephalitis demonstrate either the presence of the virus or the antibody to the virus. Attempts to isolate Japanese encephalitis virus from clinical specimens are usually unsuccessful because of low titers and the rapid production of neutralizing antibody. Isolates are sometimes obtained from CSF (in which case it is associated with a failure of antibody production and a high mortality rate) (55) or from brain tissue (either at autopsy or post-mortem needle biopsy). Japanese encephalitis virus RNA has been detected in human CSF samples using reverse transcriptase polymerase chain reaction (41). However, its reliability as a routine diagnostic test has yet to be shown. Of the various serological methods available, the IgM capture ELISA has proved the most useful. The presence of anti-Japanese encephalitis virus IgM in the CSF has a sensitivity and specificity of less than 95% for CNS infection with Japanese encephalitis virus (09). The ELISA was modified to a simple “dot-blot” nitrocellulose membrane-based format for diagnosis of Japanese encephalitis for use in the rural settings where the disease often occurs (92).

World Health Organization recommended laboratory criteria for confirmation (104):

Management

There is currently no specific antiviral treatment for Japanese encephalitis, though a number of compounds have shown some efficacy against the virus in vitro or in animal models including nitric oxide, ribavirin and interferon alpha (80; 93). In a randomized, controlled trial, oral ribavirin has not been found effective in patients with Japanese encephalitis (53).

Interferon alpha-2a was tested in a double-blind placebo-controlled trial on children with Japanese encephalitis, but with negative results (80; 93; 29). Minocycline, in an experimental study, has been found to confer complete protection in mice following Japanese encephalitis infection. Following treatment with minocycline neuronal apoptosis, microglial activation, active caspase activity, proinflammatory mediators, and viral titer were markedly decreased in Japanese encephalitis-infected mice on the ninth day (67). However, a randomized trial failed to provide desired benefit in pediatric patients with acute encephalitis syndrome (50). The study included patients of Japanese encephalitis as well.

A randomized double-blind placebo-controlled trial of intravenous immunoglobulin for Japanese encephalitis was conducted in Nepal to assess the feasibility. It was observed that intravenous immunoglobulin augments the development of neutralizing antibodies in these patients. The authors conclude that intravenous immunoglobulin is an option for Japanese encephalitis treatment; however, therapeutic potential of intravenous immunoglobulin needs to be investigated (78). Chen and colleagues assessed intravenous immunoglobulin in 124 children with severe Japanese encephalitis and noted that the patients who received intravenous immunoglobulin had a shorter duration of unconsciousness and number of days of seizures, a dull response to light, the pyramidal sign, and the meningeal-irritation sign (13). In Korea, Son and colleagues treated eight patients with combination therapy that included immunoglobulin, ribavirin, and interferon alpha-2b (94). During follow-up, four treated patients (50%) showed partial recovery, whereas one patient (12.5%) showed complete recovery. Among the three patients who received supportive management only, one (33.3%) showed partial recovery, and the clinical course of the other two remained unchanged.

Patients are given supportive treatment that involves controlling convulsions with raised intracranial pressure when they occur. Corticosteroids were given for many years, but a double-blind randomized placebo-controlled trial of dexamethasone failed to show any benefit (38). Aspiration pneumonia is a common occurrence in patients with a reduced gag reflex. Careful nursing care and physiotherapy are needed to reduce the risk of bedsores, malnutrition, and contractures.

Special considerations

Pregnancy

Japanese encephalitis occurred in five pregnant women during an epidemic in India, two of whom aborted and one of who had an apparently normal child. The virus was isolated from one of the aborted fetuses (12).