A total of 194 children with features of AES were screened, and 56 (28.8%) children (57.1% boys) were diagnosed with laboratory confirmed JE during the study period. Anti-JE IgM was detected in both CSF and serum in 44 children, and in only CSF in another 12 children. Two children were below 1 year of age, 17 between 1-5 years and the rest between 5-12 years age group; median (IQR) age of study population was 6 year 3 month (5 year 10 month, 9 year 1 month). Most common clinical features were fever (n=56, 100%), altered sensorium (n=51, 91.1%), seizures (n=36, 64.3%), signs of meningeal irritation (n=27, 48.2%) and headache (n=21, 37.5%). Glasgow Coma Scale of 8 or less was observed among 12 children (21.4%) at admission. Median (IQR) duration of symptoms before admission and duration of hospitalization was 3.5 (2,5) days and 15.7 (11, 24.2) days, respectively. MRI of brain could be performed in 38 children, of which 26 (68.4%) were abnormal. Common sites of involvement were thalamus (n=22, 84.6%), basal ganglia (n=16, 61.5%), cortex (n=12, 46.2%), brainstem (n=9, 34.6%), medial temporal lobe (n=6, 23.1%) and cerebellum (n=2, 7.7%). Hemorrhagic lesion was found in 3 children (11.5%) in addition to involvement of other parts of brain; 2 had cerebral hemorrhage and 1 had subdural hemorrhage. Ten cases (17.9%) died during the hospital stay. At the time of discharge, 17 children (30.4%) had severe sequelae, 5 (8.9%) had moderate sequelae, 6 (10.7%) developed minor sequelae, and 18 children (32.1%) showed full recovery as per LOS.

At the end of 2 year 6 month of follow-up, we observed full recovery among all children with minor sequelae. Two out of 5 children categorized as moderate sequelae at the time of discharge showed full recovery; 1 child improved and had only minor sequelae. Two children with severe sequelae died within 2 weeks of discharge. Of the 15 surviving patients with severe sequelae, one improved considerably and had only minor sequelae, three improved and were categorized as moderate sequelae, and nine children remained as severe sequelae. Three fully recovered children, and two children each from moderate and severe sequelae group were lost to follow-up. At the completion of the study, it was observed that among 37 children remaining under follow-up, 23 (62.2%) had recovered fully and 14 (37.8%) were left with variable degrees of sequelae (Fig. 1).

Fig. 1 Sequelae at different stages of follow-up in Japanese encephalitis affected children.

Motor deficit was noted in 21 children (45.6%) at discharge; quadriparesis in 14, hemiparesis in 6, and monoparesis in one child. With rehabilitation therapy, satisfactory motor improvement was noted in majority of children (66.7%) within the first year of follow-up. Behavioral abnormalities evolved fully at 1 month of discharge and were noted among 24 children; predominant features were excessive anger (n=12), irritability (n=8), aggressiveness (n=5), sudden bouts of unexplained cry or laughter (n=3), and irrelevant talking (n=1). Four children were unable to recognize family members initially, and the problem persisted in one of them. At the end of follow-up, abnormal behavior persisted in 10 (27%) children.

In the acute phase of the disease, 36 children presented with seizures, mostly of generalized tonic-clonic type; 17 of them needed two or more anti-epileptic drugs (AEDs). EEG was performed in 26 children, 18 (69.2%) were abnormal. These children were having recurrent seizures and AED was continued during follow-up. Among 12 children receiving AEDs at 2 years, therapy was stopped in six as they were seizure free with normal EEG, but two children had relapse of seizures after stoppage of drugs, and therapy was restarted. Till the end of the study, 8 children (21.6%) were on AED.

Among cases under follow-up, 23 children were school-going. Poor scholastic performance was observed in 8 (21.6%) children in the long term; another 7 of them became drop-outs due to motor deficits, behavioral problems and apprehension of seizures. Dystonia was noted in 12 children (26.1%) at the time of discharge, which improved substantially within first 6-9 months. Two children showed persistence of language problem, one with motor aphasia and another with global aphasia. Feeding problems were seen predominantly in first 6 months of follow-up; mostly due to motor deficit, in-coordination and abnormal behavior.

In the present study, 56 children diagnosed with Japanese encephalitis were evaluated by Liverpool Outcome Score which showed mortality of 17.9% and 30.4% with severe sequelae at discharge. At the end of 2.5 year follow-up, 62.2% children recovered fully and 37.8% children were left with variable degree of sequelae.

Previous studies have reported a wide range of mortality (8-25%) and severe sequelae (11-25%) with Japanese encephalitis [5,6,12-15]. Subjective nature, and therefore lack of uniformity of classification, and varying duration of follow-up may be responsible for wide range of sequelae noted in different studies. The various morbidities observed in our cohort are in agreement with previous studies [5-7,13,14]. The extent of improvement among different morbidities varied in our study population. Few patients with poor clinical and radiological features showed unexpected remarkable improvement during follow-up. Whereas some survivors with severe sequelae showed improvement of different morbidities; nevertheless they could not be placed at better functional grading as some other domains did not improve. In addition, residual neuro-psychiatric problems prevented a significant proportion of children from returning to normal life.

Only few studies have described long term outcome of Japanese encephalitis affected children beyond 1 year, most probably due to remote residence of patients causing difficulty in follow-up [5-7]. Improvement of morbidities was noted mostly within initial 9-12 months of follow up, and there was no noteworthy additional improvement afterwards. Previous studies also shown majority of improvements within initial 6-12 months for most of the Japanese encephalitis survivors and neuro-logical status at initial months of discharge was predictive of long term outcome [5,6]. Some authors noted neurological deterioration (microcephaly and hyper-active behavior) several years after discharge in some survivors and suggested the need for long term follow up [6]. However, we did not observe worsening of neuro-psychiatric status in any child till the end of follow up.

Relatively smaller sample size is a limitation of the present study. Similarly, we were unable to predict which category of children might improve and to what extent. We suggest future studies to look into these aspects. Considering high mortality and long term morbidities, preventive aspects of the disease need to be prioritized.

Acknowledgements: Dr Sharmistha Bhattacherjee, Department of Community Medicine, North Bengal Medical College and Hospital, for helping with study design and statistical analysis.

Ethics clearance: Institutional Ethics Committee, North Bengal Medical College; No: PCM/2015-16/603BK, dated December 30, 2015.

Contributors: AD: conception of the study, acquisition of data and revising the manuscript for important intellectual content. SSN: design of the study, acquisition of data and drafting the manuscript; MD: acquisition of data revising the manuscript for important intellectual content; SB: interpretation of data and revising the manuscript for important intellectual content. All the authors approved the version to be published and agreed to be accountable for all aspects of the work ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.