Skip to main content
NCBI home page
UKPMC Funders Author Manuscripts logoLink to UKPMC Funders Author Manuscripts
. Author manuscript; available in PMC: 2024 Nov 1.
Published in final edited form as: J Med Virol. 2024 May 1;96(5):e29666. doi: 10.1002/jmv.29666

Diagnostic markers of acute encephalitis syndrome and COVID-associated multisystem inflammatory syndrome in children from Southern India

Tina Damodar 1,✉, Cordelia Dunai 2,3, Namratha Prabhu 1, Maria Jose 1, L Akhila 1, Uddhava V Kinhal 4, K Anusha Raj 4, Srilatha Marate 1, A V Lalitha 5, Fulton Sebastian Dsouza 6, Sushma Veeranna Sajjan 7, Vykuntaraju K Gowda 8, G V Basavaraja 8, Bhagteshwar Singh 9,10,11, P V Prathyusha 12, Kukatharmini Tharmaratnam 13, Vasanthapuram Ravi 1, Ruwanthi Kolamunnage-Dona 13, Tom Solomon 3,14,15, Lance Turtle 2,3,16, Ravi Yadav 17, Benedict D Michael 2,3,14,15, Reeta S Mani 1,✉
PMCID: PMC7616670  EMSID: EMS198927  PMID: 38738569

Abstract

Acute encephalitis syndrome (AES) in children poses a significant public health challenge in India. This study aims to explore the utility of host inflammatory mediators and neurofilament (NfL) levels in distinguishing etiologies, assessing disease severity, and predicting outcomes in AES. We assessed 12 mediators in serum (n = 58) and 11 in cerebrospinal fluid (CSF) (n = 42) from 62 children with AES due to scrub typhus, viral etiologies, and COVID-associated multisystem inflammatory syndrome (MIS-C) in Southern India. Additionally, NfL levels in serum (n = 20) and CSF (n = 18) were examined. Clinical data, including Glasgow coma scale (GCS) and Liverpool outcome scores, were recorded. Examining serum and CSF markers in the three AES etiology groups revealed notable distinctions, with scrub typhus differing significantly from viral and MIS-C causes. Viral causes had elevated serum CCL11 and CCL2 compared with scrub typhus, while MIS-C cases showed higher HGF levels than scrub typhus. However, CSF analysis showed a distinct pattern with the scrub typhus group exhibiting elevated levels of IL-1RA, IL-1β, and TNF compared with MIS-C, and lower CCL2 levels compared with the viral group. Modeling the characteristic features, we identified that age ≥3 years with serum CCL11 < 180 pg/mL effectively distinguished scrub typhus from other AES causes. Elevated serum CCL11, HGF, and IL-6:IL-10 ratio were associated with poor outcomes (p = 0.038, 0.005, 0.02). Positive CSF and serum NfL correlation, and negative GCS and serum NfL correlation were observed. Median NfL levels were higher in children with abnormal admission GCS and poor outcomes. Measuring immune mediators and brain injury markers in AES provides valuable diagnostic insights, with the potential to facilitate rapid diagnosis and prognosis. The correlation between CSF and serum NfL, along with distinctive serum cytokine profiles across various etiologies, indicates the adequacy of blood samples alone for assessment and monitoring. The association of elevated levels of CCL11, HGF, and an increased IL-6:IL-10 ratio with adverse outcomes suggests promising avenues for therapeutic exploration, warranting further investigation.

Keywords: acute encephalitis syndrome, chemokines, COVID-associated multisystem inflammatory syndrome, cytokines, inflammatory markers, neurofilament

1. Introduction

Acute encephalitis syndrome (AES) is a significant public health concern in India, particularly affecting children.1 While primarily associated with infectious etiologies, the broad definition of AES includes patients with acute fever and altered mental state, covering diverse causes such as systemic infections, metabolic derangements, or post-infectious neurological complications. Prompt identification of the etiology is crucial for initiating timely and targeted treatment, leading to positive outcomes. However, diagnosis is hindered by overlapping clinical features and limitations in common methods like IgM ELISA and pathogen-based PCRs, which yield inconclusive or negative results in about 50% of cases.1,2 IgM antibodies can persist for an extended period post-acute illness and exhibit cross-reactivity with other circulating pathogens. Delays in presentation and sampling, typical in developing countries, impede critical cerebrospinal fluid (CSF) examination, leading to unreliable PCR test results.1–4

During the 2020–21 COVID-19 outbreak, patients with scrub typhus and dengue, common contributors to AES in India, were occasionally misdiagnosed as COVID-19.5,6 Simultaneously, hospitals in India noted a rise in pediatric cases with AES-like features, later identified as multisystem inflammatory syndrome associated with COVID-19 (MIS-C).7,8 The unclear neurological spectrum in children with MIS-C9,10 complicates the ongoing challenge of identifying AES causes.

While viral central nervous system (CNS) infections involve direct invasion of resident cells leading to inflammatory responses, neurological complications in scrub typhus primarily occur through vasculitis triggered by endothelial cell invasion of bacterium Orientia tsutsugamushi.11,12 In contrast, MIS-C results from an exaggerated immune response following prior SARS-CoV-2 exposure, characterized by a cytokine storm affecting multiple organ systems, including the CNS.13 Given the differing underlying pathophysiology but similar clinical presentations, there is an urgent need for tests to distinguish significant etiologies in childhood AES presentations.

Global literature highlights the significance of cytokines and chemokines in encephalitis, suggesting their potential use as biomarkers for identifying specific causes, assessing disease severity, and predicting outcome.14,15 However, available data on inflammatory markers in the Indian population is mostly derived from studies focusing on adults infected with the Japanese encephalitis virus (JEV).14 Recent surveillance data indicates a shift in prevalent causes of AES in India, with infections like scrub typhus, dengue, and chikungunya emerging as the predominant causes.1,2,16 This shift highlights the need to investigate host inflammatory responses and potential biomarkers in cases caused by these emerging infections. Additionally, while it is known that infections such as scrub typhus and dengue, similar to MIS-C, can trigger a cytokine storm,5,6,17 the specific differences in cytokine and chemokine profiles among these various causes remain unclear.

Another potential biomarker, neurofilament light chain (NfL), is a neuron-specific protein, found in the neuronal cytoplasm. It is consistently released at low levels from axons into both CSF and blood. Increased NfL levels are associated with neurodegenerative diseases and traumatic brain injury.18 Recent literature indicates that NfL is linked to encephalopathy and unfavorable outcomes in infectious conditions like malaria, meningitis, pneumonia, and COVID-19.19–22 While the ability of elevated NfL levels to differentiate between various causes of CNS infections remains unclear,23 its diagnostic and prognostic potential in Indian children with AES is unexplored. Additionally, since these proteins are consistently released into the bloodstream and can withstand freezing and thawing,24 there is a need to explore serum NfL as a potential biomarker for different causes of AES.

This study aimed to compare clinical and laboratory characteristics, investigate serum and CSF inflammatory markers, and NfL profiles among children with AES in southern India, focusing on scrub typhus, viral etiologies, and MIS-C. It also sought to identify markers indicating disease severity and predicting patient outcomes.

2. Methods

Children aged 1 month to 18 years meeting the Indian National Vector Borne Disease Control Program and WHO case definition for AES,25 with illness duration of less than 30 days at presentation, were prospectively enrolled at three tertiary care hospitals in Bangalore, Karnataka, India—Indira Gandhi Institute of Child Health, St. John’s Medical College and Hospital, Vani Vilas Hospital, from March 2020 to December 2022. In brief, the case definition included children with fever and altered mental status (altered behavior/personality, irritability, lethargy, drowsiness, altered speech) with or without new-onset seizures. The study received approval from the institutional ethics and review boards of the hospitals and the coordinating center, National Institute of Mental Health and Neurosciences (NIMHANS). The study team, trained in obtaining consent from parents/guardians and assent from older children, followed approved procedures and forms for the process. A subset of few patients with different etiologies, as mentioned below, were selected for this study.

2.1. Microbiological tests

Utilizing a previously outlined laboratory algorithm2 (Figure S1), blood and CSF samples underwent extensive testing for infectious causes of AES. Briefly, initial assays included serological tests for JEV, dengue, chikungunya, Leptospira sp, and scrub typhus. Subsequent tests involved real-time PCR for bacterial pathogens, herpes simplex virus (HSV)−1/2, enterovirus, varicella-zoster virus (VZV), mumps virus, and parechovirus. Samples were further stored at −80°C for additional tests mentioned below, with a focus on minimizing freeze-thaw cycles.

Children with viral etiologies were categorized as “AES-Viral,” while those with Orientia tsutsugamushi as the causal pathogen were categorized as “AES-Scrub typhus.” Patients meeting the 2020 CDC MIS-C case definition without an alternative diagnosis were grouped as “AES-MIS-C.”26 The patient distribution in each etiological group was as follows: AES-Viral (n = 21), AES-Scrub typhus (n = 29), and AES-MIS-C (n = 12).

2.2. SARS-CoV-2 anti-nucleocapsid antibody and RNA detection

Quantitative determination of anti-nucleocapsid antibodies was performed in all serum samples using the Elecsys Anti-SARS-CoV-2 N electrochemiluminescence immunoassay (Roche Diagnostics International Ltd). A cutoff index ≥1.0, as recommended by the manufacturer, indicated a reactive/positive result for anti-SARS-CoV-2 antibodies. Additionally, serum and CSF specimens underwent real-time PCR for SARS-CoV-2 RNA detection. RNA extraction utilized the QIAamp Viral RNA Mini Kit, and real-time PCR was conducted using the NIV-ICMR Single Tube Four Target Assay Kit.27

2.3. Clinical findings and patient outcome

Clinical characteristics and routine laboratory findings were entered into an online platform.28 Clinical findings of children with AES-MIS-C were described based on 2020 CDC MIS-C case definition,26 which includes fever, laboratory evidence of inflammation, clinically severe illness requiring hospitalization, multisystem organ involvement (>2), and positive current or recent SARS-CoV-2 infection by RT-PCR, serology, or antigen test; or exposure to a suspected or confirmed COVID-19 case within 4 weeks before symptom onset, and absence of an alternative diagnosis. Admission Glasgow Coma Scale (GCS) scores were recorded, with a score of 15 considered normal and ≤14 deemed abnormal. Patient outcomes, recorded by telephone consultation 3 months after discharge using the Liverpool Outcome Score (LOS),29 were classified good at a score of 5 or 4 (indicating no or minor disability) and poor as scores of ≤3 (indicating severe or moderate disability or death).

2.4. Cytokine and chemokine profiling

Serum and CSF cytokines/chemokines were measured using a Bio-Plex Pro Human Cytokine Screening Panel (Bio-Rad; Cat No. 12007283). Seventeen human cytokines/chemokines were assessed, with Panel 1 including IL-1RA, IL-12(p40), CCL11 (Eotaxin), GM-CSF, CCL2 (MCP-1), GROα (CXCL1), HGF, and M-CSF, while Panel 2 included IL-1β, IL-2, IL-6, IL-10, IFN-γ, TNF, IL-8, IL-17A, and IL-1α. Briefly, samples and standards were incubated with antibody-conjugated beads, biotinylated detection antibodies, and streptavidin–phycoerythrin. The plate was read using a Bio-Plex MAGPIX multiplex reader (Bio-Rad), and data were analyzed using Bio-Plex Manager Software version 6.0. Only mediators detected in >80% of samples were included in analyses to avoid bias due to undetectable levels or missing data.15 The balance between pro-inflammatory and anti-inflammatory cytokines was assessed by calculating the IL-1β to IL-1RA ratio in CSF and the IL-6 to IL-10 and IL-17A to IL-10 ratios in serum.

2.5. NfL levels

NfL in CSF and serum was analyzed using a SIMOA® immunoassay (Quanterix Corporation) following a two-step procedure as described previously.30 Serum samples were added without dilution, while CSF samples underwent a 1:25 dilution on-site, followed by a 4× instrument dilution for both CSF and serum specimens, resulting in a final dilution of 1:100 for CSF and 1:25 for serum. A calibration curve with provided calibrators covered the expected concentration range, and NfL concentrations were interpolated using a 1/y^2 weighted four-parameter logistic curve fit, following the manufacturer’s instructions.

2.6. Statistical analysis

Statistical analyses and data visualization utilized R version 3.6.3 (The R Project for Statistical Computing) and Prism 8 version 10.1.0 (GraphPad Software). Descriptive statistics for categorical variables were presented as frequencies and percentages, while continuous variables were summarized using median and interquartile range (IQR) after assessing normality through visual inspection of histograms. Categorical data were compared using the Chi-square or Fisher’s exact test, depending on sample size. Continuous skewed variables among the three etiological groups were compared using the Kruskal–Wallis test with Bonferroni correction. Additionally, differences between poor and good outcomes were assessed using the Mann–Whitney U test. A p-value < 0.05 was considered statistically significant. Explanatory variables identified following initial descriptive analysis were fitted into logistic regression models to evaluate their predictive value for determining etiology. Final model selection was performed using a “criterion-based approach” to minimize the Akaike information criterion and maximize the concordance statistic (C-statistic). We further controlled for the potential confounding effects of age and duration of illness by including them as covariates in the logistic regression model.

Exploratory correlations were examined using Spearman’s correlation test for examining relationships between CSF and serum cytokine/chemokine levels, CSF and serum NfL levels, NfL and cytokine/chemokine levels, GCS and cytokines/chemokines levels, and GCS and NfL levels. Heatmaps were generated from the correlation matrices to visually represent the mediators and show those that have a strong positive (dark red) or a strong negative (dark blue) correlation, as previously described.15 A subtraction matrix heatmap was created by subtracting CSF cytokine correlations from serum cytokine correlations. The 2D cytokine network analyses were created using the Spearman’s rank correlation coefficients of greater than 0.6 to generate lines between the cytokines using the qgraph package in R software, and matrix differences were assessed by the Steiger’s test31

3. Results

Out of the 62 children in the study, 12 (19%) were classified as AES-MIS-C, 29 (47%) as AES-Scrub typhus, and 21 (34%) as AES-Viral, determined by the laboratory algorithm. The median duration between illness onset and the day of sample collection after hospitalization varied across groups, with 7 (IQR 5–9) days for AES-Scrub typhus, 5 (IQR 4–8) days for AES-Viral, and 9 (IQR 6.5–17) days for AES-MIS-C (p = 0.023). Serum samples from all 62 children underwent testing for anti-SARS-CoV-2 nucleocapsid antibodies and SARS-CoV-2 PCR. For cytokine/chemokine assays, 58 serum and 42 CSF samples from 62 children were available, while for determining NfL levels, 20 serum and 18 CSF samples from 21 children were available.

3.1. SARS-CoV-2 anti-nucleocapsid antibody and RNA detection

Anti-nucleocapsid antibodies were detected in all AES-MIS-C cases, 83% of AES-Scrub typhus cases (24/29), and 62% of AES-Viral cases (13/21). No samples tested positive for SARS-CoV-2 RNA by real-time PCR.

3.2. Clinical and routine laboratory parameters

Table S1 outlines the characteristics of children with AES-MIS-C. All 12 had fever, positive anti-nucleocapsid antibodies, and evidence of inflammation. Hematological involvement was observed in all 12 children, and cardiac involvement and features of shock in 8 (67%). Gastrointestinal, respiratory, and mucocutaneous involvement were observed in 9 (75%), 6 (50%), and 5 (42%) children, respectively. Neurological findings were nonspecific and included altered sensorium in the form of lethargy, inconsolable cry, drowsiness, abnormal speech, refusal of feed, decreased activity, irrelevant talk or irritability; seizures in 9 (75%), hypertonia, signs of meningeal irritation in 2 (16.7%) each, headache and limb weakness in one child each.

Table 1 details variations in clinical and laboratory variables among the three groups. Significant differences include age, illness and hospitalization duration, personality/behavioral changes, respiratory abnormalities, hepatomegaly, blood neutrophil and lymphocyte percentages, platelet count, direct bilirubin levels, serum albumin levels, CSF white cell count, CSF lymphocyte count, and CSF protein concentrations.

Table 1. Differences in clinical and laboratory findings between the three etiological groups.

Clinical/laboratory variables n AES-Viral (N = 21) n AES-Scrub typhus (N = 29) n AES-MIS-C (N = 12) p-Value Multiple comparison
Age (years) 21  1.0 (0.7, 7) 29   9 (6, 12) 12    2.5 (1.4, 7.5) <0.001 Scrub typhus vs MIS-C (p = 0.033) Scrub typhus vs Viral (p ≤ 0.001)
Gender Male 21  12 (57.1) 29 14 (48.3) 12 9 (75) 0.29
Female    9 (42.9) 15 (51.7) 3 (25)
Glasgow coma scale score (at admission) 15   11 (8, 15) 23    13 (11, 15) 10    8.5 (8, 12.3) 0.047 -
Duration of illness (days) 21 3.0 (2, 4) 29 5 (4, 7) 12          7 (3.8, 10.5) 0.001 Viral vs. Scrub typhus p = 0.001; Viral vs MIS-C p = 0.0331
Admission in ICU required No 21    8 (38.1) 29 13 (44.8) 12 3 (25) 0.49
Yes   13 (61.9) 16 (55.2) 9 (75)
Change in personality/behavior No 21 13 (65) 29 29 (100) 12 12 (100) <0.001
Yes  7 (35) 0 (0) 0 (0)
Abnormal speech No 21 15 (71.4) 29 25 (86.2) 12 11 (91.7) 0.31
Yes   6 (28.6)   4 (13.8)  1 (8.3)
Irrelevant talk No 21 19 (90.5) 29 27 (93.1) 12 11 (91.7) 1
Yes 2 (9.5) 2 (6.9) 1 (8.3)
Seizures No 21   5 (23.8) 29 15 (51.7) 12 3 (25) 0.08
Yes 16 (76.2) 14 (48.3) 9 (75)
Respiratory symptoms (cough/shortness of breath/sore throat) No 21 16 (76.2) 29 27 (93.1) 12   4 (33.3) <0.001
Yes   5 (23.8) 2 (6.9)   8 (66.7)
Musculoskeletal symptoms (joint/muscle pain) No 21 18 (85.7) 29 28 (96.6) 12 12 (100) 0.24
Yes   3 (14.3) 1 (3.4) 0 (0)
Gastrointestinal symptoms (diarrhea/pain abdomen/abdominal distension) No 21 13 (61.9) 29 15 (51.7) 12    5 (41.7) 0.52
Yes   8 (38.1) 14 (48.3)    7 (58.3)
Lymphadenopathy No 21  21 (100) 29 24 (82.8) 12 12 (100) 0.07
Yes 0 (0)   5 (17.2) 0 (0)
Edema No 21  19 (90.5) 29 22 (75.9) 12 11 (91.7) 0.35
Yes   2 (9.5)  7 (24.1) 1 (8.3)
Mucocutaneous findings (Conjunctival congestion/Skin rash/Eschar) No 21  17 (81) 29 22 (75.9) 12   7 (58.3) 0.35
Yes   4 (19)  7 (24.1)   5 (41.7)
Hepatomegaly No 21 17 (81) 29 13 (44.8) 12 6 (50) 0.03
Yes   4 (19) 16 (55.2) 6 (50)
Splenomegaly No 21  21 (100) 29 24 (82.8) 12 12 (100) 0.07
Yes 0 (0)  5 (17.2) 0 (0)
Cranial nerve abnormality No 21  21 (100) 29 28 (96.6) 12 12 (100) 1.000
Yes 0 (0.0) 1 (3.4) 0 (0.0)
Cerebellar signs No 21  20 (95.2) 29 25 (86.2) 12 12 (100) 0.411
Yes 1 (4.8)  4 (13.8) 0 (0.0)
Meningeal irritation No 21 15 (71.4) 29 14 (48.3) 12 10 (83.3) 0.088
Yes 6 (28.6) 15 (51.7)   2 (16.7)
Total leukocyte count (× 109/L) 21      12.1 (7.7, 16.1) 29  12.8 (8.3, 15.5) 12    14.3 (9.2, 17.5) 0.756 -
Neutrophils % 21      63.8 (42.2, 73.7) 29  54.8 (43.8, 69) 12      74.3 (55.4, 83.4) 0.022 Scrub typhus vs MIS-C (p = 0.01)
Lymphocytes % 21      27.5 (19.1, 46.7) 29      38.4 (27.4, 51.91) 2      21.1 (12.1, 37.1) 0.010 Scrub typhus vs MIS-C (p = 0.011)
Platelets (× 109/L) 21      196 (79, 376.5) 29     108 (36, 156.5) 12        159 (51.3, 267.5) 0.019 Scrub typhus vs Viral (p = 0.016)
Direct bilirubin (mg/dL) 17   0.2 (0.1, 0.4) 25   0.6 (0.2, 2.1) 10   0.2 (0.1, 0.6) 0.021 Scrub typhus vs Viral (p = 0.023)
Aspartate transaminase (AST) (IU/L) 17     61.6 (24.9, 286.4) 28   144.4 (79.7, 397.1) 10     127.1 (36.6, 265.5) 0.170 -
Alanine transaminase (ALT) (IU/L) 17    44.5 (18.3, 148.1) 28    79.6 (54.2, 145.6) 9         80 (24.5, 274.2) 0.304 -
Serum albumin (g/dL) 16 3.6 (3.1, 4.2) 27 2.7 (2.3, 3.2) 10 3.1 (2.7, 4.1) 0.006 Scrub typhus vs Viral (p = 0.005)
Urea (mg/dl) 17 28.5 (20, 37.2) 26   27.2 (22.1, 31.71) 11  37.7 (20.9, 67.7) 0.199 -
Creatinine (mg/dL) 21   0.3 (0.3, 0.5) 29   0.4 (0.3, 0.5) 12 0.5 (0.4, 0.7) 0.165 -
CSF total white cell count (cells/μL) 14    3 (0, 88.3) 20      36.5 (10.8, 82.51) 0 2 (0, 6.8) 0.006 Scrub typhus vs MIS-C (p = 0.006)
CSF lymphocyte count (cells/μL) 14    3 (0, 83.3) 20 27 (10, 56) 9 2 (0, 8.5) 0.017 Scrub typhus vs MIS-C (p = 0.019)
CSF neutrophil count (cells/μL) 14   0 (0, 4.5) 20               1.5 (0, 19) 8            0 (0, 0) 0.104 -
CSF protein concentration (mg/dL) 14 23.4 (18, 52.9) 20       64.8 (48.3, 97.81) 10      21 (10.5, 39.3) 0.001 Scrub typhus vs MIS-C (p = 0.002) Scrub typhus vs Viral (p = 0.009)
Duration of hospitalization (days) 15   9 (7, 9.5) 25 8 (5, 9) 12       17 (11.8, 28.5) 0.002 Viral vs. MIS-C p = 0.049; Scrub typhus vs. MISC p = 0.002
Open in a new tab

Note: Categorical variables are represented as no. of patients (%) and continuous variables as Median (Q1, Q3); N = Total no. of patients in each etiology, n = no. of patients in whom the variable was analyzed. Statistically significant p-values are marked in bold.

Abbreviations: AES, Acute encephalitis syndrome; CSF, cerebrospinal fluid; MIS-C, multisystem inflammatory syndrome associated with COVID-19.

The median (IQR) age of the AES-Scrub typhus group was 9 (6, 12) years, while it was 2.5 (1.4, 7.5) years for AES-MIS-C and 1.0 (0.7, 7) years for AES-Viral. Multivariable logistic regression revealed that higher age predicted AES-Scrub typhus over AES-Viral [Odds Ratio (CI, p-value): 1.52 (1.18–2.23, p = 0.006)] and AES-MIS-C [Odds Ratio (CI): 1.37 (1.13–1.7, p = 0.005)]. Longer duration of illness was an independent predictor for AES-MIS-C over AES-Viral [Odds Ratio (CI): 1.26 (1.05–1.61, p = 0.028)]. A substantial majority (93%) of AES-Scrub typhus cases were aged ≥3 years (27/29) compared with 45% in AES-Viral/MIS-C (15/33) (p < 0.005). Assessing the diagnostic performance of age ≥3 years in distinguishing AES-Scrub typhus from AES-Viral/MIS-C, indicated a sensitivity (95% CI) of 93% (77.2%–99.1%), specificity of 54.5% (36.3%–71.9%), positive predictive value of 64.3% (55.0%–72.6%), negative predictive value of 90% (69.5%–97.3%), and an overall accuracy of 62.6% (59.8%–83.1%). Therefore, while age ≥3 years may provide an initial indication, it is essential to note that further diagnostic laboratory confirmation is necessary for a definitive diagnosis.

3.3. Inflammatory mediator profile

Analyzing cytokine/chemokine profiles, we examined 9 serum and 10 CSF samples from AES-MIS-C, 28 serum and 17 CSF samples from AES-Scrub typhus, and 21 serum and 15 CSF samples from AES-Viral. Mediators not identified in >80% samples (CSF & Serum IL-12p40, GM-CSF, IL-2, IL-1α; CSF IL-10, IL-17A; and Serum IL-1β) were excluded from analysis. The cytokine/chemokine profiles of children in the three etiological groups are shown in Tables 2 and 3 and represented in Figure 1. Serum CCL11, CCL2, and HGF concentrations significantly differed between etiological. Logistic regression was fitted to predict AES-Viral considering serum CCL11 and CCL2 as predictors. The results indicated higher serum CCL11 as predictive of AES of viral etiology over scrub typhus [AES Viral versus AES-Scrub typhus: 1.013 (1.003–1.023, p = 0.010)]. In CSF, AES-Scrub typhus exhibited significantly higher levels of IL-1RA (p = 0.012), IL-1β (p = 0.001), and TNF (p = 0.001) than AES-MIS-C (Table 3), while higher CCL2 levels were predictive of AES-Viral over AES-Scrub typhus [AES-Viral vs AES-Scrub typhus: Odds Ratio (CI, p-value): 1.02 (1.004–1.036, p = 0.012)] using logistic regression. After accounting for age and duration of illness as potential confounders, the observed relationships between cytokine levels etiological groups were consistent with the initial findings.

Table 2. Serum inflammatory mediator profiles of patients in the study.

Mediator (pg/mL) AES-Viral (n = 21) AES-Scrub typhus (n = 28) AES-MIS-C (n = 9) Multiple comparisons
Median (Q1, Q 3) p-Value
IL-1RA   2601.3 (796.1, 5323.2)        3049 (748.7, 9104.3)      3569 (731, 9783.5) 0.898 -
CCL11   166 (113.6, 261)            77.1 (43, 114.1)    185.1 (71.1, 239.5) 0.001 Scrub typhus vs Viral (p = 0.001)
CCL2   181.8 (123.1, 463.7)       60 (31.1, 135.4)    132.2 (96.9, 312.1) 0.004 Scrub typhus vs Viral (p = 0.003)
GROα     546.5 (339, 948.5)  230.2 (53.8, 765.9)    281.1 (53.8, 557.7) 0.080 -
HGF      1304 (471.5, 3635.6)    817.1 (342.6, 1430)       2489.4 (1503.1, 3858.6) 0.018 Scrub typhus vs MIS-C (p = 0.025)
M-CSF   44 (19.9, 99.5)  38.8 (15.4, 57.3) 19.8 (16, 36.6) 0.223 -
IL-6 8.8 (4.1, 24.5)   8.9 (2.1, 21.1) 12.2 (1.4, 44.1) 0.955 -
IL-10 9.6 (1.7, 31.6) 10.2 (3.5, 27.3)  3.8 (1.9, 15.7) 0.526 -
IFN-γ 1 (0.4, 6)    1 (0.4, 7.8) 0.6 (0.4, 1.4) 0.489 -
TNF 10.8 (6.1, 35.9) 10.5 (4.6, 27.8) 12.9 (5.8, 20.2) 0.902 -
IL-8     31.5 (11.8, 223.5)   36.8 (9.5, 122.9)    19.4 (11.8, 157.9) 0.952 -
IL-17A 0.7 (0.5, 4.9) 0.7 (0.3, 7.5) 1.8 (0.3, 6.3) 0.983 -
IL-6/IL-10          0.9 (0.5, 6) 0.7 (0.3, 1.2) 2.4 (0.3, 3.8) 0.396 -
IL-17A/IL10 0.2 (0.1, 0.5) 0.2 (0.1, 0.4) 0.1 (0.1, 0.7) 0.754 -
Open in a new tab

Note: Statistically significant p-values are marked in bold.

Abbreviations: AES, Acute encephalitis syndrome; CSF, cerebrospinal fluid; MIS-C, multisystem inflammatory syndrome associated with COVID-19.

Table 3. CSF inflammatory mediator profiles of patients in the study.

Mediator (pg/mL) AES-viral AES-Scrub typhus AES-MIS-C Multiple comparisons
n Median (Q1, Q3) n Median (Q1, Q3) n Median (Q1, Q3) p-Value
IL-1RA 15     214.9 (63.8, 2085.2) 17      1345.5 (919.9, 1925.7) 10    70.9 (17.2, 838.2) 0.012 Scrub typhus vs MIS-C (p = 0.01)
CCL11 15 1.6 (0.7, 2.6) 17   1.7 (1.1, 3.7) 10 1.8 (0.2, 4.1) 0.691 -
CCL2 15     151 (56.6, 544.1) 17   42.4 (13, 88.5) 10     76.6 (48.5, 399.2) 0.002 Scrub typhus vs Viral (p = 0.01)
GROα 15   13.5 (13.5, 13.5) 16       13.5 (13.5, 106.7) 10    13.5 (13.5, 78.2) 0.866 -
HGF 15  224.5 (106, 860.9) 17         540 (191.7, 737.6) 10    203.6 (88.3, 1399) 0.560 -
M-CSF 15 6.5 (1.8, 8.6) 17     4 (3.5, 6.4) 10   3.5 (2.1, 5.6) 0.545 -
IL-1β 15 0.1 (0.1, 0.2) 17    0.4 (0.2, 1.1) 11   0.1 (0.1, 0.1) 0.001 Scrub typhus vs MIS-C (p = 0.001)
IL-6 15  17.3 (0.9, 105.2) 17   11.4 (3.4, 27.9) 11     4.9 (2.2, 45.7) 0.838 -
IFN-γ 15 1.1 (1.1, 2.9) 17 1.6 (0.2, 2) 11   1.1 (0.2, 1.1) 0.076 -
TNF 15 4.2 (1.1, 34.6) 17    40.2 (19.2, 51.1) 11 1.8 (1, 9.9) 0.001 Scrub typhus vs MIS-C (p = 0.001)
IL-8 15     21.3 (11, 42.3) 17   71.3 (21.5, 118) 11   29 (8.3, 63) 0.145 -
Open in a new tab

Note: Statistically significant p-values are marked in bold.

Abbreviations: AES, Acute encephalitis syndrome; CSF, cerebrospinal fluid; MIS-C, multisystem inflammatory syndrome associated with COVID-19.

Figure 1. Graphical representation of cytokines/chemokines significantly associated with etiologies and patient outcomes.

Figure 1

Open in a new tab

(A)–(G) Concentrations of mediators associated with etiological groups of AES in serum and CSF. (H)–(J) Serum concentrations of mediators associated with outcomes. AES, acute encephalitis syndrome; CSF, cerebrospinal fluid.

Around 89% (25/28) of AES-Scrub typhus cases had serum CCL11 levels <180 pg/mL compared with 50% in AES-Viral/MIS-C (15/30) (p < 0.001). A combined criteria of age ≥3 years and serum CCL11 levels <180 pg/mL was observed in 24/28 AES-Scrub typhus cases and 5/30 AES-Viral/MIS-C cases. The combined criteria led to an improved discrimination, achieving a sensitivity (95% CI) of 85.7% (67.3–95.9%), specificity of 83.3% (65.3–94.4%), positive predictive value of 82.8% (68–91.5%), negative predictive value of 86.2% (71.3–94%), and an overall accuracy of 84.5% (72.6–92.7%) for a diagnosis of scrub typhus. Similarly, a combined criteria of age ≥3 years with serum CCL2 levels <140 for a diagnosis of scrub typhus, demonstrated sensitivity and specificity of 75% and 77%, respectively and a combined criteria of age ≥3 years with CSF CCL2 < 100 demonstrated sensitivity and specificity of 82% and 56%, respectively.

Heat-map (Figure 2) and 2D network analyses (Figure S2) revealed distinct patterns among different etiologies and between CSF and serum within each etiology. Figure S3 shows subtraction heatmaps comparing the three etiologies. AES-Viral group demonstrated a coordinated upregulation of most mediators primarily in CSF, AES-Scrub typhus exhibited this phenomenon mainly in serum, and AES-MIS-C displayed a mixed pattern. The results of Steiger’s test indicated a statistically significant difference between the 2D network analyses of cytokines/chemokines in CSF of patients with AES-Scrub typhus and AES-Viral groups (Table S2).

Figure 2. Heatmaps of correlations between different mediators in serum and cerebrospinal fluid (CSF) of etiological groups.

Figure 2

Open in a new tab

*Subtraction matrix shows CSF cytokine correlations subtracted from the serum cytokine correlations with white showing the correlations which are similar between the two samples, while red and blue show the many distinct relationships (Figure 2 far-right column).

3.4. NfL levels

NfL testing was conducted on a total of 38 samples (CSF = 18, serum = 20) obtained from 21 patients. Paired CSF and serum samples were analyzed for 17 patients (AES-Viral, n = 6; AES-Scrub typhus, n = 6; AES-MIS-C, n = 5), while additional four AES-MIS-C patients were specifically examined for either serum NfL (n = 3) or CSF NfL (n = 1). The median (IQR) serum and CSF NfL values were 60.5 (23.2–153.0) and 22.0 (13.0–40.8) pg/mL, respectively.

Overall, a significant correlation was present between CSF & serum NfL (r = 0.49, p = 0.04). However, no significant differences were found between serum and CSF NfL values of different groups (Table S3). A significant negative correlation was observed between CSF Nfl and serum CCL11 concentrations (r = −0.6, p = 0.01), and a weak correlation was observed between serum NfL and HGF concentrations (r = 0.5, p = 0.03).

3.5. Glasgow coma scale score

GCS scores at admission were available for 48 (77%) children, and information on abnormal (<15) or normal15 GCS, based on clinical presentation at admission, were available for 61 (98%) children. AES-MIS-C children had significantly lower GCS scores (p = 0.04) (Table 2), with 83% having a abnormal admission GCS, compared with 67% and 62% in AES-Viral and AES-Scrub typhus, respectively. Median CSF and serum NfL levels were higher in children with abnormal GCS, but the difference was not significant (Table S4). A weak correlation existed between GCS score and serum NfL (r = −0.57, p = 0.022). In children with AES-Viral, GCS significantly correlated negatively with CSF IL-8 (r = −0.76, p = 0.01); whereas in children with AES-MIS-C, GCS significantly correlated negatively with Serum IL-6/IL-10 ratio (r = −0.88, p = 0.006).

3.6. Patient outcomes

Fifty-four children survived to hospital discharge. LOS could be assessed at 3 months after discharge in 51/62 (82%) children, with 16 (31.4%) experiencing poor outcomes and 35 (68.6%) achieving good outcomes. Full recovery (LOS = 5/5) occurred in 33 (65%) children, mild sequelae (LOS = 4/5) in 2 (4%), moderate sequelae (LOS = 3/5) in 7 (14%), and severe sequelae (LOS = 2/5) in 1 (2%). Eight children succumbed to their illness (LOS = 1/5). AES-Viral exhibited a higher percentage of poor outcomes (38%) compared to AES-Scrub typhus (17%) and AES-MIS-C (25%) (p = 0.5). Significantly higher concentrations of serum CCL11, HGF, and IL-6/IL-10 ratio were associated with poor outcomes (p = 0.038, 0.005, and 0.02, respectively) (Table 4). No significant associations were observed between CSF cytokine levels and patient outcomes. Median CSF and serum NfL values were higher in patients with poor outcomes, but these differences were not significant.

Table 4. Serum mediator profiles and NfL levels in patients with poor and good outcomes.

Mediators & NfL (pg/mL) N (total) Poor outcome N (%) = 16 (31.4) Good outcome N (%) = 35 (68.6) Total N = 51 p
IL-1RA 48 (94.1)     4419.9 (749.4–14088.5)       2601.2 (798.0–6699.1)      3341.2 (758.3–7131.7) 0.470
CCL11 48 (94.1)  185.4 (89.3–275.4)      98.7 (54.2–185.1)   113.6 (64.9–219.1) 0.038
CCL2 48 (94.1)   181.8 (119.9–427.3)     115.6 (61.4–213.1)    134.9 (67.1–269.2) 0.122
HGF 48 (94.1)     2489.4 (1253.0–6317.7)        858.7 (415.4–1610.2)      1168.3 (475.1–2497.0) 0.005
M-CSF 48 (94.1)   52.5 (21.8–81.6)     40.0 (18.1–56.7)   40.1 (19.4–61.5) 0.281
IL-6 48 (94.1) 12.4 (5.4–44.8)  8.8 (2.3–16.9)  9.2 (3.3–22.7) 0.164
IL-10 48 (94.1)  9.1 (2.4–28.9)   9.9 (2.8–18.6)  9.4 (2.6–27.2) 0.991
IFN-γ 45 (88.2) 0.6 (0.5–2.1) 1.1 (0.4–6.3) 0.9 (0.5–5.9) 0.455
TNF 47 (92.2) 10.0 (5.6–31.4)  14.2 (6.6–25.6) 11.0 (6.5–27.2) 0.825
IL-8 48 (94.1)    38.7 (15.4–146.9)    26.3 (11.6–137.3)     33.7 (12.9–139.7) 0.664
IL-17A 48 (94.1) 0.5 (0.3–0.9)                  1.1 (0.5–6.8) 0.7 (0.5–4.4) 0.139
IL-6/IL-10 48 (94.1) 2.4 (0.9–5.9)  0.6 (0.3 to 1.2) 0.9 (0.4–2.4) 0.020
IL-17a/IL-10 48 (94.1) 0.1 (0.0–0.6) 0.2 (0.1– 0.4) 0.1 (0.1–0.4) 0.311
NfL 18 (35.3)    176.5 (113.5–799.5)   54.0 (22.8–150.5)   79.0 (22.8–155.0) 0.339
Open in a new tab

Note: Statistically significant p-values are marked in bold.

Abbreviations: CSF, cerebrospinal fluid; NfL, neurofilament.

4. Discussion

This study identified distinct clinical and laboratory features, along with inflammatory mediator and NfL profiles in children with AES due to scrub typhus, viral etiologies, and MIS-C in southern India.

A significant proportion of children with AES-Scrub typhus and AES-Viral in our study exhibited positive results for SARS-CoV-2 anti-nucleocapsid antibodies, despite no prior vaccination against SARS-CoV-2. This observation implies potential prior exposure or asymptomatic infection, which presents a diagnostic challenge, as noted in previous studies.32,33 Therefore to ensure appropriate treatment strategies, an exclusion of endemic causes of AES is paramount before making a diagnosis of MIS-C in patients with neurological manifestations, especially in regions with a high prevalence of AES.6

As shown in Table 1, there were a few significant differences in the clinical and routine laboratory parameters of children with AES-MISC compared to other groups. Notably, the AES-MIS-C group had highest anti-nucleocapsid antibody cut-off [median (Q1–Q3): 68.6 (46.9–137.4)] compared with AES-Scrub typhus [35.2 (9.1–126.1)] and AES-Viral [59 (17.6–121.7)]. No AES-MIS-C case tested PCR positive for SARS-CoV-2, most likely due to delayed hospital presentation and sampling [Median (Q1, Q3): 9 (6.5, 17) days].

While reports, mostly based on studies from developed nations, suggest a median age of 9 years for MIS-C, our study, alongside multicentre studies in Asia and Africa, observed a lower median age.34–36 We hypothesize that exposure to a diverse range of pathogens and antigens through natural infections and vaccination at a younger age in these regions may foster the development of a robust immunological memory. This enhanced immune memory could potentially lead to more prompt and potent immune responses upon subsequent encounters with SARS-CoV-2, thereby contributing to MIS-C at a younger age. However, further research is warranted to validate this hypothesis and elucidate the underlying mechanisms. In contrast to AES-MIS-C, children with AES- Scrub typhus had a higher median age, as reported in other studies.37 An older age [Median (IQR): 9 (6, 12) years] was identified as an independent predictor for children with AES-Scrub typhus, distinguishing them from AES-Viral [Median (IQR): 1.0 (0.7, 7) years] and AES-MIS-C [Median (IQR): 2.5(1.4, 7.5) years]. This association suggests that higher outdoor activities and increased exposure to vectors in older age may contribute to susceptibility. Given that scrub typhus is the primary cause of childhood AES in Southern India,2 these insights provide valuable information on associated risk factors.

The cytokine profiles in our study provided distinct discriminatory information for each AES etiology. Notably, serum and CSF patterns within each etiology showed differences, highlighting the unique immune profile of the CNS compared to peripheral blood. Correlation among cytokines/chemokines within the CSF network differed significantly between scrub typhus and viral etiologies, indicating distinct underlying pathophysiological mechanisms within CNS associated with these two causes of AES (Figure 2 and Figure S2)

AES-Viral group showed the highest concentrations of both serum and CSF CCL2, followed by AES-MIS-C and AES-Scrub typhus. Elevated CCL2 levels were identified as an independent predictor and potential biomarker for differentiating AES-Viral from AES-Scrub typhus. This aligns with established knowledge that infected astrocytes, macrophages/microglia, and neurons produce elevated CCL2 levels during viral encephalitis and neuro-COVID, contributing to blood-brain barrier disruption and neuroinvasion15,38,39

Elevated CCL11 levels, linked to diminished hippocampal neurogenesis and cognitive symptoms observed in long COVID (“COVID fog”),40,41 were observed in children with AES-MIS-C. Serum CCL11, also associated with neuroinflammation and neuro-degeneration in chronic traumatic encephalopathy,42 was identified as an independent predictor and potential biomarker for distinguishing AES-Viral from AES-Scrub typhus.

A significant finding was that combining age ≥3 years with serum CCL11 < 180 pg/mL in children with AES demonstrated robust diagnostic accuracy in distinguishing AES-Scrub typhus from AES-Viral and AES-MIS-C groups. However, given the limited number of patients in the study, these findings need validation on a larger population. Additionally, increased CSF TNF was linked to AES-Scrub typhus, consistent with TNF’s recognized role in enhancing the permeability of infected brain endothelial cells in vitro and its active expression during scrub typhus infection in vivo.43,44

Interestingly, serum CCL11 and HGF, as well as the ratio of pro-inflammatory IL-6 to anti-inflammatory IL-10, exhibited significant associations with adverse outcomes across all etiologies. The ratio of CSF IL-1β to IL1RA, previously linked to poorer outcomes in HSV-1 encephalitis, underscores the essential role of balancing pro-inflammatory and anti-inflammatory signaling in pathogen control and inflammation regulation.15 Moreover, both HGF and IL-6 have been identified to correlate with severity in COVID-19.45,46 HGF, recognized for its involvement in tissue repair and brain development, may suggest a reparative response with increased expression.47 IL-6, crucial in antiviral and antibacterial immune responses, has also shown negative effects on neurogenesis in both in vitro and in vivo settings.48,49

The evaluation of NfL in CSF and sera was limited to a small number of children in each group. Consequently, while NfL levels were lower in AES-Viral compared with the other two groups, this difference did not reach statistical significance. These findings align with the results reported by van Zeggeren et al., where CSF NfL levels were lower in patients with viral CNS infections compared with those with bacterial infections, CNS inflammatory disease or systemic infection.23 Children with more severe neurological impairment (indicated by lower GCS scores) and poorer clinical outcomes had higher levels of NfL in their CSF and serum. Furthermore, as the GCS scores decreased, serum NfL levels tended to increase, though the strength of this correlation was relatively weak. These findings indicate the utility of NfL as a potential marker for assessing brain injury and predicting outcomes in children with AES, consistent with observations in other CNS conditions.23,38,50

Another significant positive correlation, albeit weak, was observed between CSF and serum levels of NfL, similar to what has been reported in patients with VZV CNS infections.51 This suggests the potential use of serum NfL as a surrogate marker for evaluating neuronal damage without the need for invasive CSF sapling. The weak negative correlation between CSF NfL and serum CCL11 may indicate that the extent of neuroaxonal damage does not precisely align with the level of neuroinflammation marked by CCL11 alone. in contrast, serum HGF, with known neuroprotective effects,47 exhibited a weak positive correlation with serum NfL. This observation suggests a concurrent biological response involving the upregulation of HGF, potentially aimed at repair and regeneration, in response to increased levels of serum NfL, a marker of neuronal damage. Interestingly, HGF was also significantly associated with poor outcomes across all etiological groups and was particularly elevated in children with MIS-C.

Our study is limited by the small sample size for testing mediator and NfL profiles within each etiological group and the lack of samples from healthy controls. By grouping viral causes of AES into a single group, potential differences among specific viral etiologies may have been overlooked. Furthermore, since all participants were enrolled during the COVID-19 pandemic, most had positive SARS-CoV-2 antibodies, which complicated definitive diagnosis. Additionally, there were significant differences in sampling times in the disease process and average age between these patient groups, which may have influenced the measured levels of inflammatory markers in the study. However, even after controlling for these two factors, results of the regression model remained consistent. These limitations highlight the importance of validating these findings in a larger, age-matched population study to enhance the generalizability and reliability of the results.

In summary, viral AES is associated with increased serum and CSF levels of CCL2 and CCL11, indicating their potential as biomarkers that warrant further investigation. The composite criterion—children aged ≥3 years with serum CCL11 < 180 pg/mL—shows promise in differentiating scrub typhus from other causes, necessitating validation in a larger population. Elevated HGF levels, positively correlating with brain injury markers, were observed in children with poor outcomes and MIS-C. The association of elevated CCL11, HGF, and IL-6:IL-10 ratio with poor outcomes highlights potential therapeutic strategies that warrant further investigation. While serum NfL analysis holds promise as an alternative to CSF NfL for assessing neuronal injury in AES, its efficacy in distinguishing etiologies, severity, and outcomes requires further exploration.

Supplementary Material

Supporting Information

Acknowledgments

The authors acknowledge the technical assistance provided by Dr Sarada Subramanian and Ms Geethu Krishna in performing the NfL assay, and Dr Monojit Debnath for support in performing the luminex assay. This work was supported by grants from the Indian Council of Medical Research (ICMR) to Reeta S. Mani (Project ID: 2021-3668) and DBT/Wellcome Trust India Alliance Fellowship IA/E/15/1/503960 awarded to Tina Damodar. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Funding information

Indian Council of Medical Research (ICMR) (Project ID: 2021-3668); DBT/Wellcome Trust India Alliance Fellowship IA/E/15/1/503960

Footnotes

Author Contributions

Tina Damodar contributed to funding acquisition, conceptualization, data curation, methodology, formal analysis, investigation, and writing the original draft. Cordelia Dunai participated in formal analysis and contributed to writing the original draft. Namratha Prabhu, Maria Jose, Srilatha Marate, and Akhila L were involved in investigation, methodology, and project administration. Uddhava V Kinhal, Anusha Raj K, A V Lalitha, Fulton Sebastian Dsouza, Sushma Veeranna Sajjan, Vykuntaraju K Gowda, and G V Basavaraj were responsible for patient recruitment for the study. Prathyusha P V, Ruwanthi Kolamunnage-Dona and Kukatharmini Tharmaratnam conducted formal analysis. Tom Solomon, Lance Turtle, Vasanthapuram Ravi, and Ravi Yadav provided supervision and reviewed the manuscript. Benedict D Michael contributed to conceptualization, supervision, writing the original draft, and review & editing. Reeta S. Mani was involved in conceptualization, methodology, writing the original draft, review & editing, funding acquisition, and supervision. All authors revised the manuscript, approved it for publication, and agreed to be accountable for all aspects of the work.

Conflict of Interest Statement

C. D. and B. D. M. are supported to conduct COVID-19 neuroscience research by the UKRI/MRC (MR/V03605X/1) and by the NIHR Health Protection Research Unit (HPRU) in Emerging and Zoonotic Infections at the University of Liverpool; B. D. M. is also supported for additional neurological inflammation research due to viral infection by grants from: the NIHR [award CO-CIN-01], the Medical Research Council [MC_PC_19059] the MRC/UKRI (MR/V007181/1), MRC (MR/T028750/1), Wellcome (ISSF201902/3) and Medical Research Foundation (MRF) [MRF-CPP-R2-2022-100003]. L. T. is supported by the National Institute for Health Research Health Protection Research Unit (NIHR HPRU) in Emerging and Zoonotic Infections (NIHR200907) at the University of Liverpool in partnership with Public Health England (PHE), in collaboration with Liverpool School of Tropical Medicine and the University of Oxford. L. T. is based at the University of Liverpool. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, the Department of Health or Public Health England. L. T. has received consulting fees from MHRA; and from AstraZeneca and Synairgen, paid to the University of Liverpool; speakers’ fees from Eisai Ltd, and support for conference attendance from AstraZeneca.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available due to privacy or ethical restrictions.

References

  • 1.Ravi V, Hameed SKS, Desai A, et al. An algorithmic approach to identifying the aetiology of acute encephalitis syndrome in India: results of a 4-year enhanced surveillance study. Lancet Global Health. 2022;10(5):e685–e693. doi: 10.1016/S2214-109X(22)00079-1. [DOI] [PubMed] [Google Scholar]
  • 2.Damodar T, Singh B, Prabhu N, et al. Association of scrub typhus in children with acute encephalitis syndrome and meningoencephalitis, Southern India. Emerging Infect Dis. 2023;29(4):711–722. doi: 10.3201/eid2904.221157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Ellul M, Solomon T. Acute encephalitis – diagnosis and management. Clin Med. 2018;18(2):155–159. doi: 10.7861/clinmedicine.18-2-155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Murhekar MV. Acute encephalitis syndrome and scrub typhus in India. Emerging Infect Dis. 2017;23(8):1434. doi: 10.3201/eid2308.162028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Farhat SK, Srivastava PK, Baskar P, Krishnan J. Inflammatory cytokines in scrub typhus and COVID-19. Indian J Public Health. 2023;67(1):184. doi: 10.4103/ijph.ijph_873_22. [DOI] [PubMed] [Google Scholar]
  • 6.Dhooria GS, Kakkar S, Pooni PA, et al. Comparison of clinical features and outcome of dengue fever and multisystem inflammatory syndrome in children associated with COVID-19 (MIS-C) Indian Pediatr. 2021;58(10):951–954. doi: 10.1007/s13312-021-2329-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Rastogi S, Gala F, Kulkarni S, Gavali V. Neurological and neuroradiological patterns with COVID-19 infection in children: a single institutional study. Indian J Radiol Imaging. 2022;32(4):510–522. doi: 10.1055/s-0042-1755250. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Chintha L, Magar S, Vaidya V, Bhartiya S, Mehta K. Neurological manifestations of multisystem inflammatory syndrome in children associated with COVID-19 in a tertiary care centre. Int J Contemp Pediatr. 2023;10(3):372–375. [Google Scholar]
  • 9.Bova SM, Serafini L, Capetti P, et al. Neurological involvement in multisystem inflammatory syndrome in children: clinical, electroencephalographic and magnetic resonance imaging peculiarities and therapeutic implications. An Italian single-center experience. Front Pediatr. 2022;10(10):932208. doi: 10.3389/fped.2022.932208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Abbati G, Attaianese F, Rosati A, Indolfi G, Trapani S. Neurological involvement in children with COVID-19 and MIS-C: a retrospective study conducted for more than two years in a pediatric hospital. Children. 2022;9(12):1809. doi: 10.3390/children9121809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Dahm T, Rudolph H, Schwerk C, Schroten H, Tenenbaum T. neuroinvasion and inflammation in viral central nervous system infections. Mediators Inflamm. 2016;2016:8562805. doi: 10.1155/2016/8562805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Garg D, Manesh A. Neurological facets of scrub typhus: a comprehensive narrative review. Ann Indian Acad Neurol. 2021;24(6):849–864. doi: 10.4103/aian.aian_739_21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Sharma C, Ganigara M, Galeotti C, et al. Multisystem inflammatory syndrome in children and Kawasaki disease: a critical comparison. Nat Rev Rheumatol. 2021;17(12):731–748. doi: 10.1038/s41584-021-00709-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Soltani Khaboushan A, Pahlevan-Fallahy MT, Shobeiri P, Teixeira AL, Rezaei N. Cytokines and chemokines profile in encephalitis patients: a meta-analysis. PLoS One. 2022;17(9):e0273920. doi: 10.1371/journal.pone.0273920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Michael BD, Griffiths MJ, Granerod J, et al. Characteristic cytokine and chemokine profiles in encephalitis of infectious, immune-mediated, and unknown aetiology. PLoS One. 2016;11(1):e0146288. doi: 10.1371/journal.pone.0146288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Jain P, Prakash S, Tripathi PK, et al. Emergence of orientia tsutsugamushi as an important cause of acute encephalitis syndrome in India. PLoS Neglected Trop Dis. 2018;12(3):e0006346. doi: 10.1371/journal.pntd.0006346. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Nayak S, Panda PC, Biswal B, et al. Eastern India collaboration on multisystem inflammatory syndrome in children (EICOMISC): a multicenter observational study of 134 cases. Front Pediatr. 2022;10:834039. doi: 10.3389/fped.2022.834039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Gaetani L, Blennow K, Calabresi P, Di Filippo M, Parnetti L, Zetterberg H. Neurofilament light chain as a biomarker in neurological disorders. J Neurol Neurosurg Psychiatry. 2019;90(8):870–881. doi: 10.1136/jnnp-2018-320106. [DOI] [PubMed] [Google Scholar]
  • 19.Balanza N, Francis CK, Crowley VM, et al. Neurofilament light chain as a biomarker of neuronal damage in children with malaria. J Infect Dis. 2024;229(1):183–188. doi: 10.1093/infdis/jiad373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Ehler J, Petzold A, Wittstock M, et al. The prognostic value of neurofilament levels in patients with sepsis-associated encephalopathy—a prospective, pilot observational study. PLoS One. 2019;14(1):e0211184. doi: 10.1371/journal.pone.0211184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Chung HY, Wickel J, Oswald M, et al. Neurofilament light chain levels predict encephalopathy and outcome in community-acquired pneumonia. Ann Clin Transl Neurol. 2023;10(2):204–212. doi: 10.1002/acn3.51711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Chekrouni N, van Soest TM, Brouwer MC, Willemse EA, Teunissen CE, van de Beek D. CSF neurofilament light chain concentrations predict outcome in bacterial meningitis. Neurol: Neuroimmunol Neuroinflamm. 2021;9(1):e1123. doi: 10.1212/NXI.0000000000001123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.van Zeggeren IE, Ter Horst L, Heijst H, Teunissen CE, van de Beek D, Brouwer MC. Neurofilament light chain in central nervous system infections: a prospective study of diagnostic accuracy. Sci Rep. 2022;12(1):14140. doi: 10.1038/s41598-022-17643-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Altmann P, Leutmezer F, Zach H, et al. Serum neurofilament light chain withstands delayed freezing and repeated thawing. Sci Rep. 2020;10(1):19982. doi: 10.1038/s41598-020-77098-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Government of India. Guidelines for surveillance of acute encephalitis syndrome (with special reference to Japanese encephalitis) 2006. Nov, [Accessed Aug 25, 2023]. http://nvbdcp.gov.in/WriteReadData/l892s/AES_guidelines.pdf .
  • 26.Melgar M, Lee EH, Miller AD, et al. Council of state and territorial Epidemiologists/CDC surveillance case definition for multisystem inflammatory syndrome in children associated with SARS-CoV-2 infection—United States. MMWR Recomm Rep. 2022;71(4):1–14. doi: 10.15585/mmwr.rr7104a1. https://stacks.cdc.gov/view/cdc/121710 . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Ravindra A, Mishra B, Rath S, et al. Comparability of the sensitivity of different real time PCR kits used in the detection of SARS CoV -2. Indian J Med Microbiol. 2021;39:S74. [Google Scholar]
  • 28.Clappia. No-Code Platform for Business Operations. [Accessed Aug 25, 2023]. [Internet] https://www.clappia.com/
  • 29.Lewthwaite P, Begum A, Ooi MH, et al. Disability after encephalitis: development and validation of a new outcome score. Bull World Health Organ. 2010;88(8):584–592. doi: 10.2471/BLT.09.071357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Le ND, Muri L, Grandgirard D, Kuhle J, Leppert D, Leib SL. Evaluation of neurofilament light chain in the cerebrospinal fluid and blood as a biomarker for neuronal damage in experimental pneumococcal meningitis. J Neuroinflammation. 2020;17:293. doi: 10.1186/s12974-020-01966-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Masi A, Breen EJ, Alvares GA, et al. Cytokine levels and associations with symptom severity in male and female children with autism spectrum disorder. Mol Autism. 2017;8:63. doi: 10.1186/s13229-017-0176-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Sood M, Singh R, Bhardwaj P. Pediatric scrub typhus manifesting with multisystem inflammatory syndrome: a new cause for confusion or concern—a case series. Indian J Crit Care Med. 2022;26(6):723–727. doi: 10.5005/jp-journals-10071-24200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Gupta A, Gill A. Multisystem inflammatory syndrome in a child with scrub typhus and macrophage activation syndrome. J Trop Pediatr. 2021;67(1):fmab021. doi: 10.1093/tropej/fmab021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Mohsin SS, Abbas Q, Chowdhary D, et al. Multisystem inflammatory syndrome (MIS-C) in Pakistani children: a description of the phenotypes and comparison with historical cohorts of children with kawasaki disease and myocarditis. PLoS One. 2021;16(6):e0253625. doi: 10.1371/journal.pone.0253625. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Chinniah K, Bhimma R, Naidoo KL, et al. Multisystem inflammatory syndrome in children associated with SARS-CoV-2 infection in KwaZulu-Natal, South Africa. Pediatr Infect Dis J. 2023;42(1):e9–e14. doi: 10.1097/INF.0000000000003759. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Gupta V, Singh A, Ganju S, et al. Severity and mortality associated with COVID-19 among children hospitalised in tertiary care centres in India: a cohort study. Lancet Reg Health - Southeast Asia. 2023;13:100203. doi: 10.1016/j.lansea.2023.100203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Thangaraj JWV, Vasanthapuram R, Machado L, et al. Risk factors for acquiring scrub typhus among children in Deoria and Gorakhpur Districts, Uttar Pradesh, India, 2017. Emerging Infect Dis. 2018;24(12):2364–2367. doi: 10.3201/eid2412.180695. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Guasp M, Muñoz-Sánchez G, Martínez-Hernández E, et al. CSF biomarkers in COVID-19 associated encephalopathy and encephalitis predict long-term outcome. Front Immunol. 2022;13:86615. doi: 10.3389/fimmu.2022.866153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Terry RL, Getts DR, Deffrasnes C, van Vreden C, Campbell IL, King NJ. Inflammatory monocytes and the pathogenesis of viral encephalitis. J Neuroinflammation. 2012;9(1):270. doi: 10.1186/1742-2094-9-270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Kao J, Frankland PW. COVID fog demystified. Cell. 2022;185(14):2391–2393. doi: 10.1016/j.cell.2022.06.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Venkataramani V, Winkler F. Cognitive deficits in long COVID-19. N Engl J Med. 2022;387(19):1813–1815. doi: 10.1056/NEJMcibr2210069. [DOI] [PubMed] [Google Scholar]
  • 42.Cherry JD, Stein TD, Tripodis Y, et al. CCL11 is increased in the CNS in chronic traumatic encephalopathy but not in Alzheimer’s disease. PLoS One. 2017;12(9):e0185541. doi: 10.1371/journal.pone.0185541. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Fisher J, Card G, Soong L. Neuroinflammation associated with scrub typhus and spotted fever group rickettsioses. PLoS Neglected Trop Dis. 2020;14(10):e0008675. doi: 10.1371/journal.pntd.0008675. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Soong L, Shelite TR, Xing Y, et al. Type 1-skewed neuroinflammation and vascular damage associated with Orientia tsutsugamushi infection in mice. PLoS Neglected Trop Dis. 2017;11(7):e0005765. doi: 10.1371/journal.pntd.0005765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Perreau M, Suffiotti M, Marques-Vidal P, et al. The cytokines HGF and CXCL13 predict the severity and the mortality in COVID-19 patients. Nat Commun. 2021;12(1):4888. doi: 10.1038/s41467-021-25191-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Han H, Ma Q, Li C, et al. Profiling serum cytokines in COVID-19 patients reveals IL-6 and IL-10 are disease severity predictors. Emerg Microbes Infect. 2020;9(1):1123–1130. doi: 10.1080/22221751.2020.1770129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Desole C, Gallo S, Vitacolonna A, et al. HGF and MET: from brain development to neurological disorders. Front Cell Dev Biol. 2021;9:683609. doi: 10.3389/fcell.2021.683609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Borsini A, Merrick B, Edgeworth J, et al. Neurogenesis is disrupted in human hippocampal progenitor cells upon exposure to serum samples from hospitalized COVID-19 patients with neurological symptoms. Mol Psychiatry. 2022;27(12):5049–5061. doi: 10.1038/s41380-022-01741-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Soung AL, Vanderheiden A, Nordvig AS, et al. COVID-19 induces CNS cytokine expression and loss of hippocampal neurogenesis. Brain. 2022;145(12):4193–4201. doi: 10.1093/brain/awac270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Kuhle J, Gaiottino J, Leppert D, et al. Serum neurofilament light chain is a biomarker of human spinal cord injury severity and outcome. J Neurol, Neurosurg Psychiatry. 2015;86(3):273–279. doi: 10.1136/jnnp-2013-307454. [DOI] [PubMed] [Google Scholar]
  • 51.Tyrberg T, Nilsson S, Blennow K, Zetterberg H, Grahn A. Serum and cerebrospinal fluid neurofilament light chain in patients with central nervous system infections caused by varicella-zoster virus. J Neurovirol. 2020;26:719–726. doi: 10.1007/s13365-020-00889-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
EMS198927-supplement-Supporting_Information.zip (2.5MB, zip)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available due to privacy or ethical restrictions.

RESOURCES