A Comparison of Clinical and Radiological Findings in Adults and Children With Japanese Encephalitis FREE

JAPANESE ENCEPHALITIS (JE) is the most common human endemic encephalitis, affecting more than 50 000 individuals and resulting in more than 10 000 deaths per year. In newly endemic areas, both adults and children are affected. In areas where JE infection has been endemic for several years, mainly children are affected and adults are not, owing to herd immunity. In regions where JE has been controlled by immunization for a long time, such as Japan and Korea, the virus may affect elderly individuals. In India, JE was first reported in 1956 in Vellore.¹ The virus gradually extended north, and cases of JE have been reported throughout India. Both adults and children have been affected.

In a multivariate analysis of the prognostic predictors for adults and children with JE, the best predictors of outcome were age, Glasgow Coma Scale score, and reflex changes.² Poor prognostic predictors reported for children included the presence of high fever, deep coma, a short prodromal stage, abnormalities in tone and breathing, decerebration, and a focal neurological deficit.^{3 - 5}

The structural, functional, and immunological differences between the brains of adults and children, ongoing myelination, and neuronal plasticity may influence the clinical pictures, radiological findings, and the outcome of patients with JE. In the available literature, there are no comparative studies evaluating the clinical laboratory findings and outcome in adults and children with JE. In this study, we analyzed the clinical and radiological features of JE in adults and children.

This study was based on a retrospective analysis of patients with JE diagnosed and treated during the past 10 years. Diagnosis of JE was determined by virus-specific antibody titers, polymerase chain reaction, or virus isolation. The antibody titers were assessed using the following methods: hemagglutination inhibition (HI) titer for the JE virus infection in acute and convalescent serum samples of patients with JE or a 2-mercaptoethanol test in the serum,^{6 - 7} IgM capture through an enzyme-linked immunosorbent assay,⁸ isolation of the JE virus,⁷ and polymerase chain reaction.⁹ Patients with acute encephalitic syndrome and positive results for 1 or more of the previous tests were included in this study. The choice of method was based on the availability of the test at that particular time. The HI titer was used until 1998 (9 children and 23 adults); the remaining 25 patients were tested using 1 of the remaining 3 methods alone or in combination with the HI titer. The tests were performed in cerebrospinal fluid except for the HI titer and 2-mercaptoethanol tests, which were assessed in serum samples. Acute encephalitic syndrome refers to fever with altered sensorium without a diagnosis of malaria or bacterial meningitis.

The patients underwent a detailed clinical evaluation. Consciousness was assessed by the Glasgow Coma Scale and muscle power by the 0-V Medical Research Council scale.¹⁰ Muscle tone was categorized as reduced, normal, or increased and was further classified by spasticity or rigidity. Deep tendon reflexes were classified as normal, reduced, or exaggerated. Presence of focal reflex loss with exaggerated or normal response in another area was categorized according to mixed pattern. Depending on the patient's consciousness level, the sensations were tested for pinprick, joint position, vibration, and cortical sensations. Extrapyramidal signs such as rigidity, dystonia, dyskinesia, tremor, and other movement disorders were noted.

A cranial computed tomographic scan was performed using a high-resolution computed tomographic scanner, and 10-mm axial sections were obtained parallel to the orbitomeatal plane. Cranial magnetic resonance imaging was performed using a 2-T scanner operating at 1.5 T (Magnetome SP; Siemens, Munich, Germany). T1-weighted (repetition time [TR]/echo time [TE]/excitation = 500 ms/50 ms/3 ms), proton density (TR/TE/excitation = 2000-2500 ms/15-20 ms/1 ms), and T2-weighted (TR/TE/excitation = 2000-2500 ms/80-90 ms/1 ms) spin-echo sequences were obtained. Electroencephalography (EEG) was performed using a 10- to 18-channel EEG with a 10-20 system of electrode placement. Sagittal, temporal, and transverse montages were recorded using bipolar and referential derivations. Response to intermittent photic stimulation varying from 3 to 30 Hz, hyperventilation, and eye opening were evaluated in the patients who could cooperate. A 30-minute EEG record was obtained for all patients. The EEG was analyzed for background activity, any right-to-left asymmetry, rhythmic activity, and epileptiform discharges.¹¹

Cerebrospinal fluid was examined in all patients for protein, glucose, cells, bacteria, and fungi. Outcome was defined at the end of 6 months as poor (bedridden), partial (needing help with daily activities), or complete recovery (able to perform daily activities independently).¹² Patients who were younger than 15 years were included in the pediatric group; those older than 15 years were included in the adult group. Various clinical, radiological, and laboratory parameters in children and adults were compared using the χ² test.

During the last 10 years, we treated 67 patients with virologically diagnosed JE of whom 30 were children and 37 were adults. These patients came from the nearby Indian districts of Uttar Pradesh (87%), Bihar (5.2%), and Madhya Pradesh (2.6%) and Nepal (5.2%).

In the pediatric group, a positive result was obtained for the HI titer in 18 patients, the IgM enzyme-linked immunosorbent assay in 17, the polymerase chain reaction in 6, and virus isolation in 1. In the adult group, positive findings occurred for the HI titer in 21 patients, the IgM enzyme-linked immunosorbent assay in 5, and the polymerase chain reaction in 8. The positivity of these tests was not statistically different between the 2 groups (χ²₃ = 7.30; P = .06). The mean age of the children was 9.5 years; the mean age of the adults was 35.5 years.

Seizures were present in 23 adults (62.2%) and in 17 children (56.7%), and the difference was not statistically significant. Most seizures were generalized tonic-clonic, although focal seizures with secondary generalization were present in 6 children and 7 adults. The seizures were generally infrequent and easily controlled with phenytoin sodium or carbamazepine monotherapy except in a 2-year-old patient with status epilepticus. At the 6-month follow-up visit, none of the patients had had recurrence of seizures.

The depth of coma was comparable in adults and children. Focal neurological deficit was present in 19 children (63.3%) and 20 adults (54.1%). Quadriplegia was the most common characteristic, present in 28 patients (71.8%), followed by hemiplegia, which was present in 11 patients (28.2%). Features of anterior horn cell involvement as evidenced by focal reflex loss, localized wasting, and fibrillations on the electromyogram were present in 25 adults (67.6%) and 17 children (56.7%).

Movement disorders were common in both adults and children; however, the pattern and frequency of the movement disorders differed between the 2 groups. Dystonia was more common in children (20 [66.7%] of the children vs 7 [18.9%] of the adults), whereas parkinsonian features were frequent in both groups (22 [59.5%] of the adults and 21 [70%] of children).

The EEG results were abnormal in all of the children; they revealed a slowing of background activity in all patients and spike-wave discharges in 4 patients. In adults, the EEG results were abnormal in 34 patients, which included 4 patients with alpha coma; 29 patients had a slowing of background activity. Of these patients, 1 patient each had associated right-to-left asymmetry and spike-wave discharges. Cranial magnetic resonance imaging results were more frequently abnormal than computed tomographic results, and the abnormalities were similar in both adults and children (Figure 1 and Figure 2). The only difference was the presence of cysticercosis in 3 children; none of the adults had this infection. The patients with cysticercosis did not have a history of seizure before developing JE, and all had partial seizures during the acute encephalitic phase.

At 6 months, 6 adults died, whereas none of the children did. Surviving adults, however, had better outcomes compared with the children. Details of the clinical parameters in adults and children are presented in Table 1, and the radiological changes are listed in Table 2.

Comparison of the clinical parameters in adults and children revealed that dystonia (χ²₁ = 13.78; P<.001) was more frequent in children and the 6-month outcome (χ²₃ = 9.70; P = .02) was worse in children compared with adults. The pattern of abnormalities in EEG results (χ²₃ = 8.71; P = .03) was also significantly different between the 2 groups.

Our study reveals some important differences between adults and children with JE. Although the adult patients with JE had a higher mortality rate in the acute stage, their 6-month outcome was better compared with the children's. The children, on the other hand, had a higher frequency of dystonia. The pattern of abnormalities in EEG results was also different between the 2 groups.

Our results differ from those in the published literature in several ways. Children with JE reportedly have more severe encephalitis as evidenced by deep coma, high frequency of seizures, and higher mortality and morbidity rates. In the literature, seizures have been reported to be more common in children compared with adults; only 10% of adult patients have seizures compared with 64% to 80% of children.^{4 ,13 - 14} A recent study from Vietnam,⁵ which included mainly children, also reported a high frequency of seizures. The higher frequency of seizures in the children from the Vietnamese study could be owing to a difference in the herd immunity as well as referral bias. In the Vietnamese study, 24 (62%) of 40 patients with seizures died or had severe sequelae.⁵ In a study of 65 patients with JE, which included 19 children, 11 had seizures, but seizure was not associated with poor prognosis. Diagnosis of JE in this study was, however, based on epidemiological, clinical, radiological, or virological criteria.¹⁵ In the present study, none of the children died. Seizures were not a serious problem and did not correlate with sequelae. Coinfection of cysticercosis and JE in the same patient was reported and was attributed to the role of pigs in the life cycle of both JE and cysticercosis; pigs are the amplifying host in JE and the intermediate host in cysticercosis. Three children with JE in our study also had neurocysticercosis, and all these children had partial seizures during the acute stage of encephalitis. A high frequency of association between JE and neurocysticercosis was reported in an autopsy study from southern India,¹⁶ which exceeded the chance association. Coinfection of JE and cysticercosis was also reported in 8 of 26 brain autopsy results from China.³ The pathological changes in patients with JE were more prominent in the patients who had neurocysticercosis. Cysticercosis may alter the blood-brain barrier and may account for central nervous system invasion by the JE virus. The high frequency of seizures and poor prognosis in patients with JE found by Solomon et al⁵ raises questions about other underlying causes. In the absence of computed tomographic scans and magnetic resonance imaging, the association of cysticercosis or other infection with JE needs further evaluation before attributing it to the higher frequency of seizures in patients with JE.

A wide variety of movement disorders have been reported in patients with JE, including parkinsonian features, myoclonic jerk, ocular deviations, oculogyric crisis, orofacial dyskinesia, mouth-open dystonia, tongue dystonia, torticollis, and teeth-clenching dystonia.^{14 ,17} In another study, movement disorders were found in 10.5% of patients with JE. During the acute phase of the illness, the movement disorders included head nodding, coarse tremor, choreoathetosis, dystonia, and parkinsonian features.¹⁸ Movement disorders were also reported as late sequelae of JE; of 68 patients with JE examined 5 years after the acute phase of the illness, the movement disorders included monotonous speech in 7 patients; masked facies in 11 patients; tremor in 7 patients; chorea, tics, and athetosis in 7 patients; and parkinsonian syndrome in 8 patients.¹⁹ In our study, parkinsonian features were prominent in adults, whereas both dystonia and parkinsonian features were prominent in children. The difference was statistically significant (P<.001). Dystonia is more common in children following hypoxic-toxic-metabolic insult and is attributed to the vulnerability of the developing central nervous system. Dystonia is more refractory and has been associated with a worse prognosis.^{20 - 21} A higher mortality rate in adults with JE may be owing to the inability to bear the acute stress, associated comorbidity, and immunological status. The surviving adults, however, had better outcomes compared with the children. The present study was conducted in a tertiary care teaching hospital where patients pay for their treatment, so there was a bias for patients with a high socioeconomic status or those with more serious illness. However, these variables are likely to be equally applicable to both adults and children and are unlikely to influence the conclusions of this study. The observed differences between adults and children need further evaluation.

Corresponding author: U. K. Misra, DM, Department of Neurology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Raebareily Rd, Lucknow 226014, India (e-mail: ukmisra@sgpgi.ac.in).

Accepted for publication September 13, 2003.

Author contributions: Study concept and design (Drs Kalita and Misra); acquisition of data (Drs Kalita, Misra, and Pandey); analysis and interpretation of data (Drs Kalita, Misra, Pandey, and Dhole); drafting of the manuscript (Drs Kalita and Misra); critical revision of the manuscript for important intellectual content (Drs Kalita, Misra, Pandey, and Dhole); obtained funding (Dr Misra); administrative, technical, and material support (Dr Misra).

We thank Rakesh Kumar Nigam, BS, for secretarial help.