Etiological spectrum of acute encephalitis syndrome in children in India: A systematic review and meta-analysis

Abstract

BACKGROUND

Acute encephalitis syndrome (AES) continues to be an important cause of illness and death among children in India. While Japanese encephalitis (JE) virus has traditionally been regarded as the leading cause, recent studies indicate that the range of causative organisms has widened considerably.

AIM

To estimate the pooled proportion of AES cases attributable to various common infectious etiologies among children aged 18 years or younger in India.

METHODS

A comprehensive search of PubMed/MEDLINE, EMBASE, and Scopus was conducted for studies published between January 2000 and December 2025. Studies conducted in India that included at least 25 children (≤ 18 years) with AES and reported laboratory-confirmed etiologies were considered eligible. Case reports and small case series were excluded. Quality assessment of included studies was performed using the Joanna Briggs Institute checklist. A random-effects model was used to obtain pooled estimates and their corresponding 95% confidence intervals (CI).

RESULTS

A total of 69 studies were included in the final analysis. Scrub typhus emerged as the most common cause, accounting for 20% (95%CI: 13%-27%) of AES cases, followed by JE at 13% (95%CI: 11%-15%). Enterovirus accounted for 6% of the cases (95%CI: 3%-9%). Dengue, herpes simplex virus and Epstein-Barr virus accounted for 5% of the cases each (95%CI: 3%-8%, 2%-9%, and 1%-11%, respectively). Mumps and chikungunya virus accounted for 4% each (95%CI: 1%-8%, and 2%-5%, respectively).

CONCLUSION

Scrub typhus and JE are the leading identifiable etiology of pediatric AES in India. Broader diagnostic methods, including molecular techniques, may improve the diagnostic yield and facilitate early targeted therapy in children with AES.

Key Words: Encephalitis; Japanese encephalitis; Scrub typhus; Dengue; Pediatrics

Core Tip: This meta-analysis of 69 studies shows that scrub typhus and Japanese encephalitis are the leading identifiable etiologies of pediatric acute encephalitis syndrome (AES) in India. Enterovirus, dengue virus, herpes simplex virus, and Epstein-Barr virus are other etiological agents detected in the AES cases. These findings underline the need to consider a wider range of infectious etiologies in children presenting with AES. Broader diagnostic methods, including molecular techniques, may improve diagnostic yield and facilitate early targeted therapy in children with AES.

INTRODUCTION

Acute encephalitis syndrome (AES) represents a major cause of neurological morbidity and mortality among children worldwide, particularly in low- and middle-income countries[1]. The syndrome is clinically defined by the acute onset of fever accompanied by altered mental status, such as confusion, disorientation, or coma, and/or new-onset seizures, reflecting inflammation or dysfunction of the central nervous system[2]. Pediatric populations are especially vulnerable, and survivors frequently experience long-term neurological sequelae, including cognitive impairment, behavioural abnormalities, and epilepsy[3].

India has reported a substantial burden of AES for several decades, with thousands of cases documented annually through national surveillance systems[4]. Historically, the Japanese encephalitis (JE) virus has been considered the predominant etiological agent of AES in endemic regions, particularly in northern and eastern parts of the country, where ecological conditions favour mosquito-borne transmission[5]. Major outbreaks have been documented since the mid-twentieth century, and the virus has been responsible for a significant proportion of encephalitis cases during epidemic periods[6]. However, the epidemiology of encephalitis in India has evolved considerably over the past two decades[7,8].

Following the introduction of large-scale JE vaccination programs, the contribution of JE to AES cases has declined in several regions[9]. At the same time, increasing evidence suggests that a wide spectrum of other infectious agents now contribute to pediatric encephalitis in India[10]. Viral pathogens such as dengue virus, enteroviruses, influenza viruses, and human herpesvirus-6 (HHV-6) have been reported in association with encephalitis outbreaks and sporadic cases[11,12]. In addition to viral causes, bacterial and rickettsial pathogens, including Orientia tsutsugamushi (scrub typhus), are increasingly recognized as important etiological agents, which are potentially treatable with timely antimicrobial therapy[13].

The etiological distribution of pediatric encephalitis in India is further influenced by substantial geographic heterogeneity[14]. Variations in climate, vector ecology, agricultural practices, and socioeconomic conditions contribute to differences in pathogen transmission across regions[15]. For example, certain arboviral infections demonstrate strong seasonal patterns related to monsoon-driven vector proliferation, while zoonotic pathogens may be linked to agricultural exposure or rodent reservoirs[16]. Consequently, the epidemiological profile of encephalitis may differ considerably between states, complicating both surveillance and clinical management.

Despite increasing research in this field, a significant proportion of AES cases remain etiologically undiagnosed. Several factors contribute to this diagnostic gap, including limited access to advanced laboratory facilities, delayed specimen collection, and the restricted pathogen panels used in many clinical studies[17]. Recent advances in diagnostic technologies, such as multiplex polymerase chain reaction (PCR) assays and metagenomic sequencing, have improved the ability to detect multiple pathogens simultaneously and identify previously unrecognized causes of encephalitis[18]. However, these technologies remain underutilized in many settings, and findings from individual studies often remain geographically limited. Given the evolving etiological landscape of pediatric encephalitis in India and the substantial regional variability in reported pathogens, synthesizing existing evidence is essential for understanding the current epidemiology of the disease. Hence, this systematic review and meta-analysis was conducted with the objective of estimating the pooled proportion of AES cases attributable to various common infectious etiologies in children aged 18 years or younger in India.

MATERIALS AND METHODS

Search strategy

Based on a preliminary review of the available literature, the review was restricted to a selected group of pathogens that have been frequently reported in association with pediatric encephalitis in India. These included JE virus, dengue virus, scrub typhus, herpes simplex virus (HSV), enterovirus, adenovirus, cytomegalovirus (CMV), Epstein-Barr virus (EBV), varicella virus, parvovirus, mumps virus, measles virus, chikungunya virus, HHV-6, West Nile virus, and Chandipura virus.

The review was conducted in accordance with the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines[19]. However, we could not register the review protocol in the International Prospective Register of Systematic Reviews or any other systematic review registry.

A comprehensive literature search was independently carried out by two authors on December 29, 2025. The search included all relevant studies published from January 1, 2000 up to the search date. Electronic databases searched included PubMed/MEDLINE, EMBASE, and Scopus.

The search strategy was developed using a combination of Medical Subject Headings and relevant free-text keywords, including “encephalitis”, “acute encephalitis syndrome”, “acute febrile encephalopathy”, “child”, “children”, “pediatric”, and “etiology”. These terms were appropriately combined using Boolean operators (“AND” and “OR”) to enhance the sensitivity and specificity of the search, and the strategy was adapted for each database based on differences in indexing and search functionality. In addition to database searching, the reference lists of relevant review articles and eligible studies were manually examined to identify any additional publications that might have been missed during the initial search. A detailed search strategy for each database has been provided in Supplementary Table 1.

Eligibility of studies

Both prospective and retrospective clinical studies published in English language were considered eligible for inclusion in this review if they fulfilled the following criteria: The study was conducted in India; it included at least 25 children aged less than 18 years with AES; and the study reported the etiological agent responsible for AES in these children based on laboratory confirmation such as serological tests, PCR, or other validated diagnostic methods.

Studies that included both pediatric and adult patients were also considered eligible, provided that separate data were reported for the pediatric population. Studies were excluded if conducted outside India, included only adult patients, or were multicentre studies involving participants from countries other than India. Publications prior to 2000 were also excluded. Additionally, small case series and individual case reports were not included. Duplicate records and studies reporting data from the same patient population were carefully screened and removed to avoid duplicate data within the review.

Data extraction and assessment of risk of bias

All potentially eligible articles identified through the search process underwent full-text evaluation by two independent reviewers. The methodological quality of each study was assessed during this process. Relevant data were extracted from each included article after the full-text review. The extracted information included the study design, study period, healthcare setting, geographical location (state) of the study centre, and total number of patients enrolled. Clinical variables recorded included the number of children diagnosed with AES and the number of children in whom each of the selected etiological agents was identified. Information regarding the diagnostic methods used for confirmation of these etiologies was also documented.

We also noted whether the data were collected during a single outbreak in a particular region, where a single pathogen might be expected to predominate, or over an extended time period. To ensure uniformity in data collection, a standardized data extraction form prepared in advance was used. Extracted data were subsequently entered into a Microsoft Excel spreadsheet for further analysis. When disagreements arose regarding study inclusion or data extraction, the two reviewers discussed the issue and reached a consensus. Furthermore, a third independent reviewer verified the extracted data for completeness and accuracy. Care was taken throughout the process to prevent duplication of data. The methodological quality and risk of bias of the included studies were assessed using the Joanna Briggs Institute (JBI) critical appraisal checklist for studies reporting prevalence data.

Outcome measures

The outcome of interest in this review was the pooled proportion of AES cases (percentage) attributable to various common infectious etiologies in children aged < 18 years in India.

Statistical analyses

A meta-analysis was carried out using MetaXL statistical software (version 5.3). Given the expected variability among the included studies, a random-effects model was used for the pooled analysis. Appropriate descriptive statistical methods were used to summarize the collected data. Whenever feasible, pooled estimates, along with their corresponding 95% confidence intervals (CI), were calculated. Statistical heterogeneity was assessed using Higgins’ I² statistic and the Cochrane Q test; I² > 50% and a P value < 0.05 in the Cochrane Q test were considered suggestive of significant heterogeneity. Appropriate sensitivity and subgroup analyses were conducted, taking into account the risk of bias, study year, and the state in which the study was conducted, wherever feasible. Publication bias was examined using visual inspection of funnel plots for symmetry and Egger’s test if > 10 studies were pooled for a particular pathogen. Funnel plot asymmetry or Egger’s test P value < 0.10 was suggestive of significant publication bias.

RESULTS

A total of 1787 records were identified through database searches. After removing 572 duplicate records, 1215 unique records remained for screening. Titles and abstracts of these records were screened, leading to the exclusion of 1098 records that did not meet the inclusion criteria. Subsequently, 117 full-text articles were retrieved; all were successfully obtained and assessed for eligibility. Of these, 48 studies were excluded because they included only adult participants or presented combined data for adults and children without separate reporting for the pediatric population or the sample size was less than 25. Finally, 69 studies met the inclusion criteria and were included in the systematic review and meta-analysis (Figure 1).

Figure 1  PRISMA flow diagram showing study selection for the systematic review & meta-analysis.

The methodological quality of the included studies was assessed using the JBI Critical Appraisal Checklist for Studies Reporting Prevalence Data. An appropriate sampling frame, appropriate sampling method, sufficient data coverage, and standard and reliable measurement of the condition were each reported in 68 studies. Adequate sample size was observed in 48 studies. However, valid methods for identifying the condition were reported in only 25 studies, which represents the main methodological limitation. This was primarily because most studies tested for only a limited number of pathogens, while other potential causative agents were not adequately investigated. An adequate response rate was reported in 67 studies.

The majority of the included studies were published after 2010. Studies reporting scrub typhus were almost entirely from 2016 onwards. In studies published after 2010, there was greater use of advanced diagnostic techniques such as multiplex PCR, which enabled identification of other pathogens, including EBV, enterovirus, and parvovirus. Similarly, infections such as measles, mumps, and chikungunya virus were predominantly reported in studies published after 2010. Chandipura virus was predominantly reported in outbreak-based studies, where it was identified as a causative agent in a significant proportion of cases. However, most studies that assessed prevalence over a longer period did not report this. Geographically, the highest number of studies was conducted in Uttar Pradesh, indicating regional clustering of available evidence. Characteristics of included studies have been described in Table 1.

Table 1 Characteristics of included studies in the meta-analysis.

Ref.	State of study	Sample size (for pediatric age group)	Predominant pathogen	Serological methods used	Molecular methods used
Tsomu et al[3], 2026	Himachal Pradesh	69	Scrub typhus	Yes	No
Abdulkader et al[2], 2025	Multiple states of North and South India	34501	JE	Yes	No
Bhardwaj et al[11], 2025	Uttar Pradesh	557	Scrub typhus	Yes	Yes
Damodar et al[30], 2025	Karnataka	587	Scrub typhus	Yes	Yes
Sidharth et al[31], 2025	Kerala	48	HHV-6, Adenovirus	Yes	Yes
Thomas et al[8], 2025	Kerala	204	HHV-6	No	Yes
Mishra et al[32], 2025	Uttar Pradesh	264	Scrub typhus	Yes	Yes
Verma et al[5], 2025	Uttar Pradesh	4116	JE	Yes	No
Kumar et al[33], 2025	Bihar	2195	JE	Yes	No
Dwibedi et al[4], 2024	Odisha and Uttar Pradesh	867	HSV	Yes	Yes
Alam et al[34], 2024	Uttar Pradesh	204	Dengue	Yes	No
Sharawat et al[35], 2024	Uttarakhand	100	Dengue	Yes	Yes
Bhardwaj et al[36], 2024	Uttar Pradesh	100	Scrub typhus	Yes	Yes
Tripathy et al[37], 2019	Odisha	834	HSV	Yes	Yes
Kakoti et al[9], 2020	Assam	140	JE	Yes	No
Gupta et al[38], 2018	Gujarat	90	HSV	Yes	Yes
Ravi et al[39], 2022	Multi-centric	11295	Scrub typhus	Yes	Yes
Tiwari et al[40], 2017	Rajasthan	283	HSV	Yes	Yes
Arankalle et al[41], 2019	Uttar Pradesh	78	Scrub typhus	Yes	Yes
Damodar et al[42], 2023	Karnataka	376	Scrub typhus	Yes	Yes
Saxena et al[43], 2009	Uttar Pradesh	36	JE	Yes	Yes
Medhi et al[44], 2017	Assam	500	JE	Yes	No
Jana et al[13], 2024	West Bengal	75	Scrub typhus	Yes	No
Baidya et al[45], 2022	Pondicherry	127	Scrub typhus	Yes	Yes
Tandel et al[46], 2019	Madhya Pradesh	75	Only reported incidence of West Nile virus infection, not the predominant pathogen	No	Yes
Swami et al[47], 2008	Chandigarh	27	JE	Yes	Yes
Singh et al[48], 2014	Uttar Pradesh	1264	Dengue	Yes	Yes
Singh et al[49], 2019	Manipur	526	JE	Yes	No
Mishra et al[22], 2024	Uttar Pradesh	130	JE	Yes	Yes
Kumar et al[50], 2008	Uttar Pradesh	265	JE	Yes	Yes
Gurav et al[51], 2010	Maharashtra	78	Chandipura virus (outbreak)	Yes	Yes
Singh et al[52], 2015	Uttar Pradesh	114	JE	Yes	Yes
Kakoti et al[21], 2013	Assam	223	JE	Yes	No
Karmarkar et al[14], 2008	Delhi	151	Enterovirus	Yes	No
Bhardwaj et al[53], 2024	Uttar Pradesh	238	Scrub typhus, Chikungunya	Yes	Yes
Rebecca et al[54], 2024	Tamil Nadu	250	Scrub typhus	Yes	Yes
Mittal et al[24], 2017	Uttar Pradesh	46	Scrub typhus	Yes	No
Tandale et al[15], 2022	Maharashtra & Telangana	198	JE	Yes	Yes
Kumar et al[55], 2023	Madhya Pradesh	110	JE	Yes	No
Tandale et al[17], 2021	Maharashtra & Telangana	140	JE	Yes	Yes
Anukumar et al[56], 2014	Kerala	69	West Nile virus (outbreak)	Yes	Yes
Sonowal et al[23], 2024	Assam	1389	JE	Yes	Yes
Adarsha et al[57], 2023	Chandigarh	31	Scrub typhus, HSV	Yes	Yes
Kabilan et al[58], 2000	Tamil Nadu	37	JE	Yes	No
Jain et al[25], 2014	Uttar Pradesh	1045	JE	Yes	No
Lewthwaite et al[59], 2010	Karnataka	232	JE	Yes	Yes
Kumar et al[60], 2011	Uttar Pradesh	90	Enterovirus (outbreak)	No	Yes
Kumar et al[61], 2018	Uttar Pradesh	100	HSV	Yes	Yes
Saxena et al[62], 2009	Uttar Pradesh	38	JE	Yes	Yes
Ranjan et al[63], 2014	Uttar Pradesh	8453	JE	Yes	Yes
Singh et al[64], 2016	Uttar Pradesh	128	Enterovirus	Yes	Yes
Bhatt et al[65], 2012	Uttar Pradesh	100	JE	Yes	No
Kumar et al[66], 2006	Uttar Pradesh	223	JE	Yes	No
Aggarwal[67], 2022	Delhi	100	JE	Yes	Yes
Kabilan et al[68], 2004	Tamil Nadu	58	JE	Yes	Yes
Balachandran et al[20], 2025	Gujarat	81	Chandipura virus (outbreak)	Yes	Yes
Kumar et al[69], 2012	Uttar Pradesh	204	Enterovirus	Yes	Yes
Beig et al[27], 2010	Uttar Pradesh	87	Enterovirus	Yes	No
Chakraborty et al[70], 2021	Assam	1185	JE	Yes	No
Thomas et al[18], 2024	Kerala	157	Enterovirus	Yes	Yes
Rathore et al[71], 2019	Odisha	820	HSV	Yes	Yes
Tandale et al[72], 2008	Andhra Pradesh (Telangana now)	52	Chandipura virus	Yes	Yes
Patgiri et al[73], 2014	Assam	144	JE	Yes	No
Kumar et al[74], 2009	Uttar Pradesh	577	JE	Yes	No
Itihas et al[75], 2023	Maharashtra	56	JE	Yes	No
Mittal et al[76], 2018	Uttar Pradesh	407	Scrub typhus	Yes	Yes
Tarai et al[77], 2019	New Delhi	157	Enterovirus	No	Yes
Narasimha Rao et al[78], 2008	Andhra Pradesh	104	Chandipura (outbreak)	Yes	Yes
Alam et al[26], 2020	Uttar Pradesh	352	Scrub typhus	No	Yes

JE: Japanese encephalitis; HSV: Herpes simplex virus; HHV-6: Human herpesvirus 6.

Scrub typhus

The pooled analysis showed that scrub typhus accounted for 20% of AES cases (95%CI: 13%-27%, I² = 99%) based on 23 studies (12332 participants) (Figure 2A). After excluding studies with at least one item rated as “No” on the JBI critical appraisal checklist, the sensitivity analysis revealed that scrub typhus was responsible for 13% of AES cases (95%CI: 6%-21%, I² = 99%). Studies conducted in Uttar Pradesh showed a pooled proportion of scrub typhus-related AES cases of 30% (95%CI: 17%-44%, I² = 98%), in contrast to studies conducted in other states, where it was 12% (6%-20%, I² = 98%).

Figure 2 Forest plot. A: Showing pooled estimate for percentage of acute encephalitis syndrome (AES) cases in children attributable to Scrub typhus; B: Showing pooled estimate for percentage of AES cases in children attributable to Japanese encephalitis; C: Showing pooled estimate for percentage of AES cases in children attributable to Dengue virus; D: Showing pooled estimate for percentage of AES cases in children attributable to enterovirus; E: Showing pooled estimate for percentage of AES cases in children attributable to herpes simplex virus. AES: Acute encephalitis syndrome; CI: Confidence interval.

Viral etiologies

JE represented 13% of AES cases (95%CI: 11%-15%, I² = 98%) across 55 studies (74512 participants) (Figure 2B). After excluding studies with at least one item rated “No” on the JBI critical appraisal checklist, the sensitivity analysis revealed that JE accounted for 7% of AES cases (95%CI: 4%-10%, I² = 98%). While studies conducted in Uttar Pradesh showed the proportion of JE cases was 13% (95%CI: 10%-16%, I² = 97%), studies conducted in other states showed the proportion of JE cases was 13% (10%-15%, I² = 98%). Proportion of AES cases caused by JE was 23% (13%-35%, I² = 96%) pooled across studies published before 2010, while proportion of AES cases caused by JE was 10% (9%-12%, I² = 98%) pooled across studies published in or after 2010.

Dengue contributed to 5% of cases (95%CI: 3%-8%, I² = 98%) across 28 studies (16145 participants) (Figure 2C). After excluding studies with at least one item rated “No” on the JBI critical appraisal checklist, sensitivity analysis revealed dengue constituted 4% of AES cases (95%CI: 2%-6%, I² = 96%). While studies conducted in Uttar Pradesh showed the proportion of dengue cases was 7% (95%CI: 3%-12%, I² = 95%), studies conducted in other states showed the proportion of dengue cases was 4% (1%-7%, I² = 98%). The proportion of AES cases caused by the dengue was 5% (1%-9%, I² = 93%), pooled across studies published before 2010, while the proportion of AES cases caused by the dengue was 5% (2%-9%, I² = 98%), pooled across studies published in or after 2010.

Enterovirus and HSV were responsible for 6% (95%CI: 3%-9%, I² = 97%) across 20 studies (10191 participants) and 5% (95%CI: 2%-9%, I² = 98%, 24 studies, 12185 participants) of AES cases, respectively (Figure 2D and E). Similarly pooled estimates for EBV, CMV, mumps, measles and parvovirus were 5% (95%CI: 1%-11%, I² = 94%, 7 studies, 1794 participants), 3% (95%CI: 1%-5%, I² = 85%, 7 studies, 1552 participants), 4% (95%CI: 1%-8%, I² = 94%, 10 studies, 2771 participants), 2% (95%CI: 1%-4%, I² = 91%, 13 studies, 5388 participants) and 3% (95%CI: 1%-7%, I² = 88%, 9 studies, 1846 participants) respectively (Supplementary Figure 1).

Pooled estimates for chikungunya virus, varicella, HHV-6, adenovirus and West Nile virus were 4% (95%CI: 2%-5%, I² = 73%, 10 studies, 3396 participants), 1% (95%CI: 1%-3%, I² = 91%, 15 studies, 5370 participants), 3% (95%CI: 1%-6%, I² = 86%, 7 studies, 1649 participants), 3% (95%CI: 1%-8%, I² = 86%, 5 studies, 1261 participants) and 1% (95%CI: 0%-3%, I² = 89%, 8 studies, 7381 participants) respectively (Supplementary Figure 2).

Chandipura virus, although not routinely identified, was reported predominantly in outbreak settings, where it accounted for 24% of AES cases (95%CI: 5%-49%, I² = 96%) based on 5 studies (455 participants) (Supplementary Figure 3), but the existing literature could not provide an accurate pooled estimate in non-outbreak settings. Assessment of publication bias using visual inspection of funnel plots did not show significant asymmetry for most pathogens (Supplementary Figures 4 and 5). This was further supported by Egger’s test (P > 0.10 for those pathogens reported in at least 10 studies) suggesting no significant publication bias.

DISCUSSION

Our review highlights the changing epidemiological landscape of pediatric encephalitis in India. Traditionally, encephalitis in the country has been strongly associated with JE virus, particularly during seasonal outbreaks in endemic regions. However, findings from the current review suggest that pediatric encephalitis should increasingly be viewed as a syndromic condition with multiple infectious etiologies rather than a disease driven by a single pathogen. This observation is consistent with recent surveillance and hospital-based studies from different parts of India, which have reported an expanding spectrum of pathogens contributing to AES[20].

Earlier studies from northern India frequently reported JE as the dominant etiology, particularly during large outbreaks[21]. In contrast, more recent investigations increasingly identify infections such as scrub typhus and dengue among children presenting with encephalitis-like illness[22]. This shift likely reflects multiple factors, including expanded JE vaccination, improved surveillance systems, and greater awareness of alternative pathogens. The growing recognition of rickettsial infections has important clinical relevance, especially in resource-limited settings where extensive diagnostic testing may not be feasible. The relatively higher proportion of treatable conditions, such as scrub typhus, suggests that early empiric therapy with doxycycline can be considered in appropriate clinical situations. At the same time, the wide range of viral causes underscores the need for a practical, stepwise diagnostic approach.

Another notable feature of pediatric encephalitis in India is the coexistence of several viral neurotropic pathogens with overlapping clinical presentations. Viruses such as HSV, enteroviruses, adenoviruses, parvovirus, mumps virus, and arboviruses can cause similar neurological manifestations, including seizures, altered sensorium, and even focal deficits. Because these infections share common clinical characteristics, distinguishing between them on clinical grounds alone is often difficult. Consequently, reliance on syndromic clinical diagnosis without adequate laboratory confirmation may lead to under recognition of certain etiologies. The presence of multiple viral pathogens with varying epidemiological patterns, therefore, complicates both clinical diagnosis and surveillance efforts[23].

However, rickettsial infections, such as scrub typhus, exhibit a somewhat different epidemiological pattern. These infections are frequently associated with rural exposure, agricultural activity, and seasonal mite proliferation. Few Indian studies have shown that central nervous system involvement in scrub typhus can closely mimic viral encephalitis, potentially delaying appropriate treatment if the possibility is not considered early. Unlike most viral encephalitides, rickettsial infections respond well to inexpensive antimicrobial agents such as doxycycline and azithromycin. This difference underscores the clinical importance of maintaining a broader differential diagnosis in children presenting with encephalitis, particularly in endemic regions[24].

The epidemiology of pediatric encephalitis in India also demonstrates considerable regional heterogeneity. Northern and eastern states historically reported the highest burden of JE, largely due to favourable ecological conditions for mosquito breeding and agricultural practices, such as rice cultivation[25]. However, recent studies suggest that the etiological profile in these regions has become increasingly diverse. In some areas, rickettsial infections and arboviral diseases have emerged as significant contributors to AES[26]. In contrast, other regions report a higher proportion of sporadic viral encephalitis caused by herpesviruses or enteroviruses. These regional variations may reflect ecological differences, vector distribution, host exposure patterns, and climatic conditions.

Apart from ecological factors, differences in health system infrastructure and diagnostic practices may also influence the reported etiological spectrum. Tertiary care hospitals often have better laboratory facilities and may therefore detect a wider range of pathogens. Conversely, smaller centres with limited diagnostic capacity may test only for a few common infections such as JE. Such variations in testing practices can lead to apparent differences in pathogen distribution across studies. Therefore, regional variation in reported etiologies should be interpreted cautiously, as it may partly reflect differences in diagnostic capacity rather than true epidemiological differences[27].

Another persistent challenge in AES research is the large proportion of cases with unidentified etiology. Even in studies using expanded diagnostic panels, a significant number of children remain without a confirmed pathogen diagnosis. Several factors may contribute to this gap. First, diagnostic investigations are often restricted to a limited set of pathogens. Second, the timing of sample collection can affect the sensitivity of molecular and serological tests. Third, cross-reactivity and limited specificity of certain serological assays may result in misclassification. Finally, a subset of cases classified under AES may represent non-infectious neurological conditions, including autoimmune encephalitis or metabolic encephalopathies[28].

The increasing availability of advanced molecular diagnostic technologies offers an opportunity to improve pathogen detection in encephalitis. Multiplex PCR panels allow simultaneous detection of multiple viral and bacterial pathogens, while metagenomic sequencing can identify previously unrecognized infectious agents. However, the use of these technologies in routine clinical practice remains limited in many regions due to cost and infrastructure constraints. Expanding access to such diagnostic tools could significantly enhance understanding of the etiological spectrum of pediatric encephalitis[29].

Several limitations should be considered while interpreting the findings of this review. The high heterogeneity observed across pooled analyses likely reflects substantial differences in geographical distribution, outbreak settings, diagnostic methods, and pathogen testing strategies across studies. A large proportion of the available studies were conducted in tertiary care hospitals. This may introduce referral bias because such centres often manage more severe cases of encephalitis. As a result, the findings may not fully represent the epidemiological situation in community or primary care settings. The geographical distribution of studies was uneven. Certain high-burden states have been studied extensively, whereas other regions of the country remain underrepresented. This imbalance may limit the ability to draw conclusions about the true national distribution of encephalitis etiologies. Considerable heterogeneity in diagnostic methods was observed across studies. Many investigations relied primarily on serological assays, which may have limitations related to cross-reactivity and variable sensitivity. In contrast, only a subset of studies employed molecular techniques such as PCR, which are generally more specific. These methodological differences may affect comparability between studies. Several studies tested only a limited number of pathogens, potentially leading to under detection of other infectious agents. In addition, some research groups contributed multiple studies from the same geographic areas, potentially leading to overrepresentation of certain regions in the available literature. Emerging or rare causes of encephalitis were not systematically evaluated in many studies. Therefore, the current evidence base may underestimate the contribution of less common pathogens. Likewise, a pooled estimate of cases with unknown etiology could not be derived due to inconsistent reporting across studies regarding the extent of pathogen testing and the number of undiagnosed cases. Furthermore, the etiological spectrum of AES may vary across pediatric age groups. However, an age-stratified analysis (e.g., infant, young children, adolescents) could not be conducted, as most included studies did not provide sufficiently detailed age-specific etiological data to permit meaningful subgroup analysis. Lastly, the protocol for this systematic review was not registered prospectively in any registry, and we included only studies published in English.

Future studies should adopt multicentre, prospective designs with uniform diagnostic protocols and broader pathogen testing to reduce variation across studies and enable more robust comparisons of results. The use of advanced molecular diagnostic techniques, along with the inclusion of community-level data, will help improve the overall quality and wider applicability of the evidence.

CONCLUSION

Scrub typhus and JE are the leading identifiable etiologies of pediatric AES in India. Broader diagnostic methods, including molecular techniques, may improve the diagnostic yield and facilitate early targeted therapy in children with AES.