Secondary Logo

Journal Logo

Original Studies

Diagnosis and Therapy of Infectious Encephalitis in Children: A Ten-Years Retrospective Study

Pata, Davide MD*,†; Buonsenso, Danilo MD‡,§,¶; Frasca, Giampiero JD*; Lazzareschi, Ilaria MD*,‡; Salerno, Gilda JD*; Turriziani Colonna, Arianna JD*; Mariotti, Paolo MD; Valentini, Piero MD*,‡

Author Information
The Pediatric Infectious Disease Journal: June 2021 - Volume 40 - Issue 6 - p 513-517
doi: 10.1097/INF.0000000000003070
  • Free


Encephalitis is a condition of brain inflammation accompanied by clinical signs of neurologic dysfunction.1 The etiology can be infectious, or noninfectious, including immune-mediated mechanisms.1

Regarding infectious encephalitis, limited comprehensive studies analyzing clinical, laboratory, microbiology, imaging, treatment, and outcomes are available in pediatric age. Recently, a few studies from Latin America and United States have been published2–4; however, viral meningoencephalitis is highly heterogeneous, varying by geographic location, and data from Europe are poor.

Consequently, we conducted a retrospective study to provide a detailed picture of children admitted with acute encephalitis in our Institution.


We conducted a retrospective analysis of children with a diagnosis of encephalitis admitted between January 1, 2008, and December 31, 2018, in our Institution, a tertiary hospital in Rome. The patients were admitted through our Emergency Department or were referred from other district general hospitals in Central Italy.

Encephalitis was defined according to the definition of the Clinical Practice Guidelines by the Infectious Diseases Society of America1 as the presence of an inflammatory process of the brain in association with clinical evidence of neurologic dysfunction. Only children whose guardian provided written informed consent were included in the analyses. The study was approved by the Institution Review Board of our Institution.

The diagnosis of encephalitis was made following the international definition of encephalitis.1 In particular, children were included if presented an altered mental status at least 24 hours and 2 of the following: fever, seizures, focal neurologic findings, cerebrospinal fluid (CSF) white cell count of ≥5 cells/mm3, abnormal brain imaging, and electroencephalogram abnormalities. In suspicion of an infectious origin, microbiologic tests such as cultures of biologic fluids (CSF, blood, urine, stool, sputum), nucleic acid amplification tests (PCR on blood and nasopharyngeal swab), and serologic tests (presence of specific IgM antibodies in acute phase and variation of the IgG antibody titer during the convalescence phase) were performed. Neuroimaging and electroencephalography (EEG) results were also reviewed to evaluate edema and hemorrhage in the temporal lobes5 or a temporal focus with periodic epileptiform discharges as seen in in herpes simplex encephalitis.6

Patients more than 18 years of age, children with immunodeficiency, chronic degenerative pathologies, postinfectious or autoimmune encephalitis, or children with evidence of bacterial/fungal CSF infection (by cultures or molecular assays) have been excluded. The differential diagnosis between autoimmune and infectious encephalitis is not simple. Encephalitis generally has a typical clinical picture: acute, monophasic, responsive to antimicrobial therapy and with prodromal symptoms in the infectious forms, while it is generally subacute, polyphasic, with psychiatric symptoms, working memory deficits, and resistance to antimicrobial therapy in autoimmune encephalitis.7 Although a rapid onset may be present in anti-NMDA receptor encephalitis, psychiatric, cognitive and speech dysfunctions, movement disturbances, and autonomic disorders are also present.7 Tests for autoimmune etiology are not part of the early evaluation of encephalitis with a characteristic clinical picture for an infectious form in our Institution. Indeed, such testing is not available in all hospitals, and results are usually available late. Furthermore, their negativity does not exclude an immune-mediated pathology. Consequently, these tests were performed only in suspected autoimmune encephalitis, excluded from this study.

The aim of this study was to provide a detailed analysis of the clinical presentation, laboratory, radiology and neurophysiology findings, microbiologic data, treatments, and outcomes of children with encephalitis. Secondary aim was to evaluate which parameters significantly affected outcome.

Data were retrospectively collected analyzing medical charts. For each patient, main epidemiologic and anamnestic findings, signs and symptoms at admission, cerebrospinal fluid analyses, details of imaging studies performed [computed tomography (CT), magnetic resonance imaging (MRI), brain ultrasound (US)], electroencephalographic results, microbiologic results, and treatment used (antibiotics, antivirals, steroids) were collected. Clinical outcome was evaluated by permanent neurologic symptoms after discharge, such as focal deficit, dysarthria, epilepsy, neurocognitive development delay (12 months after encephalitis). Patients were divided into the “good outcome” (those without neurologic sequela), and “worse outcome” (those with permanent neurologic damage or death) group.

The analysis of data included a descriptive part of the sample carried out by constructing frequency tables (absolute and percentages) for the categorical variables and with the mean ± standard deviation for the quantitative variables. The association between the dependent and independent variables has been tested using the χ2 test and the Fisher’s exact test. For all analyses, a P < 0.05 was considered significant. Analyses were performed using STATA v16.1.


Study Population

Eighty children with encephalitis-like symptoms were admitted during the study period. Of these, 24 were excluded, presenting postinfectious or autoimmune encephalitis (n. 11), acute disseminated encephalomyelitis (ADEM, n. 8), acute lymphoblastic leukemia (n. 3), and neuroblastoma (n. 2). Therefore, a total of 56 children [22 female (39.6%), mean age 4.7 years, IQR 0.7–8.7 years] with a final diagnosis of acute encephalitis were enrolled. Mean duration of hospitalization was 16 days (5–27 days). Further clinical details about the study population are reported in Table 1. Cerebrospinal Fluid Analyses (CSF) details are showed in Table 2. Microscopy and cultures of CSF were negative in all patients.

TABLE 1. - Main Clinical Data of the Study Population
Study Population (N = 56) %
Prodromal symptoms* 23 41.1
 URI 10 18
 Gastrointestinal 10 18
 URI and gastrointestinal 2 3.5
 skin rash 1 1.7
Symptoms at admission
 Fever 47 83.9
 Fatigue 14 25
 Vomiting 16 28.6
Neurologic manifestations
 Seizures 30 53.6
 Confusion 27 48.2
 Headache 18 32.1
 Loss of consciousness 15 26.8
 Meningismus 6 10.7
 Hypertonia 4 7.1
 Ataxia 3 5.4
 Respiratory failure 2 3.6
PICU admission 34 60.7
*The mean time between these manifestations and the admission at the hospital was 2.0 days.
Mean time of the PICU admission was 5 days (IQR = 1–9 days).
IQR indicates interquartile range; PICU, pediatric intensive care unit; URI, upper respiratory tract infection.

TABLE 2. - Association Between Clinical, Laboratory, and Radiologic Data and Poor Outcome
Poor Outcome n = 12 (22.4%) Good Outcome n = 41 (77.4%) OR 95% CI P
 0–4 yrs 9 (75.0%) 20 (48.8 %) 3.2 (0.74–13.34) 0.1191
 ≥5 yrs 3 (25.0%) 21 (51.2 %) 0.3 (0.07–1.34) 0.1191
 Male 7 (58.3%) 25 (61.0 %) 0.9 (0.24–3.31) 0.8693
 Female 5 (41.7%) 16 (39.0 %) 1.1 (0.30–4.13) 0.8693
Clinic presentation
 Fever 10 (83.3%) 35 (85.4 %) 0.9 (0.15–4.92) 0.8628
 Altered mental status 8 (66.7%) 24 (58.5 %) 1.4 (0.37–5.47) 0.6135
 Seizure 7 (58.3%) 22 (53.7 %) 1.2 (0.33–4.44) 0.7749
 PICU admission 10 (83.3%) 23 (56.1 %) 3.9 (0.76–20.15) 0.1027
 Pleocytosis (cells >5/mm3) 5 (41.7%) 18 (43.9 %) 0.9 (0.25–3.36) 0.8907
 Altered glycorrhachia (<40 or >70 mg/dL) 4 (33.3%) 6 (14.6 %) 2.9 (0.66–12.81) 0.1563
 Altered proteinorachie (>40 mg/dL) 4 (33.3%) 9 (22.0 %) 1.8 (0.43–7.28) 0.4238
 Altered brain MRI (performed in 46 patients) 10/12 (83.3%) 13/34 (38.2 %) 8.1 (1.52–42.84) 0.0141
 Altered EEG (performed in 47 patients) 8/9 (88.9%) 27/38 (71.1 %) 3.3 (0.36–29.24) 0.2912
CI indicates confidence intervals; OR, odds ratio; PICU, pediatric intensive care unit.


Computed tomography (CT) was performed in 36 patients (64.3%), MRI in 48 (85.7%) and, when the age of the patient allowed, cerebral ultrasound (US) in 3 (5.4%).

Thirty-three (91.7%) CT scans performed were negative. The 3 abnormal CT scans showed asymmetry of cerebral hemispheres in 1 child, diffuse hypodense areas in temporal, frontal, and occipital lobe in 2 children.

MRI was normal in 18 patients (37.5%), while the other 30 (62.5%) showed lesions (inflammatory, edematous, and vascular) localized in frontal lobes (12.5%), parietal lobes (16.7%), occipital lobes (10.4%), temporal lobes (22.9%), insula (12.5%), cerebellum (6.3%), corpus callosum (10.4%), basal ganglia (8.3%). No association was found between a specific image and an etiologic agent.

Transfontanel brain US was performed in 3 (5.4%) patients (mean age was 29 days). In 1 patient (33%), US was negative, 1 patient (33%) showed a mild cerebral edema, the third patient (33%) reported vasculopathy of lenticular-striated vessels bilaterally, related to parenchymal injury of subcortical white matter seen with MRI.

Electroencephalographic Study

Electroencephalogram (EEG) was performed in 49 patients (87.5%): 12 (24.5%) were normal, while 8 (16.3%) showed diffuse alterations on both hemispheres: irritative-type (50%), slow-type (37.5%), and background changes (12.5%). Eight EEGs showed abnormalities on a single hemisphere (half of them on the left side): they were specific (37.5%) and slow-type (62.5%). The cerebral lobes most frequently affected by EEG changes were the temporal lobe (24.5%), occipital lobe (20.4%), or both (10.2%).


Intravenous (IV) antiviral therapy with Acyclovir was used in 54 patients (96.4%), in doses of 10 mg/kg every 8 hours, doubled in children less than 3 months of age; mean duration was 10 days (3–34 days). Two patients were not treated with Acyclovir because herpetic encephalitis was initially suspected. Nine patients (16.1%) were treated with Oseltamivir for a mean length of was 5 days (5–7 days). Corticosteroids were used in 10 patients (17.9%), in 5 cases with classic dosing (intravenous methylprednisolone 0.5–1 mg/kg every 12 hours) and in 5 cases with “pulse therapy” (daily intravenous boli of methylprednisolone for 3 or 5 days followed by down-scale during 16 days). Mean duration of corticosteroid therapy was 22.5 days. The administration of corticosteroids was chosen in critically ill patients: 4 (40%) worsened during hospitalization and 6 (60%) had a radiologic pattern of cerebral edema or worsening of the cerebral lesion.


The etiologic agent was identified in 31 patients (55.4%); in the other patients, we found HSV-1 and HSV-2 (n = 11, 19.6%), Epstein-Barr Virus (EBV, n = 5, 8.9%), Influenza virus (n = 3, 5.4%), and Cytomegalovirus (CMV), Respiratory Syncytial Virus (RSV), Human Herpes Virus 6 (HHV-6), Rotavirus, Parechovirus, M. pneumonia (n = 1, 1.8% each). In 5 cases, the etiologic agent was found in the CSF, while in the others in other biologic fluids.

We enrolled 2 newborns with infectious encephalitis, 1 with HSV-1 infection and the other with Parechovirus.

Table 2 shows the main clinical, laboratory, and imaging findings in children with HSV encephalitis compared with those with other etiologies.

Children with HSV encephalitis had fever less frequently, more frequently manifested seizures (P = 0.02) and needed more frequent PICU admission (P = 0.0467), compared with children with non-HSV encephalitis.

Epstein-barr virus (EBV) and influenza were the most frequent single agents causing encephalitis after HSV. Cases of were febrile in 4 cases (80%), had seizures in 3 (60%), and needed PICU admission in 4 cases (80%). The encephalitis from influenza virus were all febrile with seizures (n = 2, 66%) and needed PICU admission (n = 3, 100%). Due to the low number of cases of children with these etiologies, statistical comparisons could not be performed.


Outcome data were available for 53 children, since 3 were lost at follow-up (Table 3). Forty-one children (77.4%) recovered without neurologic sequelae (Table 3). Eleven children (20.8%) presented neurologic sequelae, 1 child (1.9%) died, 3 (5.7%) were lost at follow-up. Among children with neurologic sequelae, the etiologic causes were to HSV (3 cases), influenza (2 cases), Epstein-Barr virus, and Parechovirus (1 case each). Abnormal brain MRI was the only factor significantly associated with a poor outcome [P = 0.01; OR, 8.1 (95% CI: 1.52–42.84)].

TABLE 3. - HSV Encephalitis
Item HSV % Other Etiology % Total OR 95% CI P
Mean age in yrs 3.5 4.9
Days of hospitalization 16 16
Female 7 63.6% 15 33.3% 22 3.5 (0.88–13.86) >0.05
Fever 7 63.6% 40 88.9% 47 0.2 (0.05–1.02) >0.05
Asthenia 2 18.2% 12 26.7% 14 0.6 (0.12–3.24) >0.05
Headache 1 9.1% 15 33.3% 16 0.2 (0.02–1.71) >0.05
Vomit 3 27.3% 15 33.3% 18 0.8 (0.17–3.24) >0.05
Seizures 10 90.9% 20 44.4% 30 12.5 (1.47–106.04) 0.0206
Meningism 0 0.0 6 13.3% 6 0.3 (0.01–5.05) >0.05
Altered consciousness 5 45.5% 29 64.4% 34 0.5 (0.12–1.75) >0.05
Gastrointestinal symptoms 2 18.2% 10 22.2% 12 0.8 (0.14–4.20) >0.05
Respiratory symptoms 1 9.1% 7 15.6% 8 0.5 (0.06–4.94) >0.05
CSF cells 4 36.4% 22 48.9% 26 0.6 (0.15–2.33) >0.05
CSF glucose 4 36.4% 7 15.6% 11 3.1 (0.71–13.48) >0.05
CSF proteins 2 18.2% 19 42.2% 21 0.3 (0.06–1.57) >0.05
Altered temperature 0 0.0 3 6.7% 3 0.5 (0.03–10.97) >0.05
Altered MRI 6 54.5% 22 48.9% 28 1.3 (0.33–4.71) >0.05
Altered EEG 8 72.7% 29 64.4% 37 1.5 (0.34–6.34) >0.05
Necessity of intensive care 10 90.9% 24 53.3% 34 8.7 (1.03–74.18) 0.0467
CI indicates confidence intervals; CSF, cerebrospinal fluid; OR, odds ratio.


Our study confirms the high heterogeneity of acute encephalitis in the pediatric population with regard to clinical, laboratory, microbiologic, and radiologic findings. In particular, etiologic diagnosis was not achieved in 31 patients (55.4%), while clinical presentations, laboratory, and instrumental data were widely variable even when we excluded with certain or probable autoimmune encephalitis. Instead, we also included patients without an etiologic agent identified because they had clinical characteristics, therapeutic responses and laboratory, radiologic, and electroencephalographic tests compatible with infectious encephalitis.

Although younger children and those admitted in PICU had more frequently neurologic deficits on follow-up, differences were not statistically significant.

In our cohort, the most frequent symptom at presentation was fever, followed by seizure and altered mental status with only 4 patients showing none of these symptoms at presentation. These findings are consistent with available literature4,8–18 and confirm difficulties in timely recognition of encephalitis in children. Importantly, in our cohort, the clinical presentations were not associated with outcome.

Our study confirmed that CSF results alone should not be used as a rule-out test in children with suspected encephalitis. In our study, we found that CSF glucose was normal or slightly elevated although in some studies in the literature it was slightly reduced.19 Forty-three patients had normal or slightly elevated CSF proteins, with the highest values of 250 mg/dL. Pleocytosis was found in 44.6% of patients. None of the CSF parameters were associated with worse outcome.

EEG was normal in almost a quarter of cases and a pathologic EEG was not associated with a worse outcome. Conversely, a study performed in Brazil found that abnormal EEG correlated with a poor prognosis,2 although it is difficult to compare different studies since EEG results may vary during disease course8 or according to pattern analysis used.20 The role of EEG need, therefore, further assessments and, although cannot be used as a rule-out test, can help the clinician.

In our series, MRI played an important role since was the only parameter significantly associated with worse outcome. This has also been seen in prior studies,2,9,21–23 our study reinforces the importance of brain MRI in children with suspected encephalitis. While MRI is not easily accessible in all settings, with CT being more available. Unfortunatley, in our series, CT was negative in most cases (91.7%). In a recent series, Sevilla-Acosta et al found that diffuse cerebral edema was the most important mortality predictor.3

Corticosteroid treatment deserves a separate discussion: it still represents a gray area in the management of infectious encephalitis. To date, the use of steroid treatment in infectious encephalitis is based on local protocol and experiences. It is usually administered for its action in situations of cerebral edema24 and to reduce inflammation.25,26 On the other hand, if used in the initial phase of viral replication, a weakening of the immune system may lead to uncontrolled viral replication and permanent damage.27

To evaluate the effect of corticosteroids in herpetic encephalitis, a multicenter, randomized, double-blind, placebo-controlled clinical trial in adult patients (GACHE) was attempted.28 However, the study was stopped due to the small number of patients recruited, with no significant differences between groups. This demonstrates the difficulty of performing clinical trials in rare diseases and the use of steroids in these conditions remains experimental.

We treated 10 patients (17.9%) with corticosteroids, and the decision to start steroid treatment was dictated more by the radiologic pattern than by the initial symptomatology or by the identified pathogen. All patients who received the cortisone had significant changes in MRI (edema, inflammation, and leukoencephalopathy). Since in our setting is difficult to obtain MRI soon after admission, steroids were started after an average of 5 days from admission. Their use did not affect outcome.

The most frequent etiologic agent was HSV (19.6%), followed by EBV (8.9%) and three cases of Influenza (5.4%). HSV children had less frequently fever (although not statistically significant, P = 0.05), developed more frequently seizures (P = 0.02) and had higher risk of PICU admission (P = 0.04), confirming a more severe clinical scenario in case of HSV children and the need of including anti-HSV treatments particularly for more severe cases.1,28,29

Our study has limitations to address. The retrospective nature is an intrinsic limitation that may have led to missing information. Second, use of steroids was based on clinician’s decisions and not based on standardized local protocols, not allowing us to properly define the role of steroids in these children. However, our study provided a large, detailed, comprehensive analyses of children with encephalitis from a European setting.

In conclusion, our study showed that encephalitis in children is a heterogeneous entity with nonspecific clinical and laboratory findings, with undefined etiologies in most times. MRI can play a primary role, both for diagnosis and prognosis. The exact role of steroids in children with encephalitis remains unclear.


1. Tunkel AR, Glaser CA, Bloch KC, et al.; Infectious Diseases Society of America. The management of encephalitis: clinical practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis. 2008;47:303–327.
2. Valle DAD, Santos MLSF, Giamberardino HIG, et al. Acute childhood viral encephalitis in Southern Brazil. Pediatr Infect Dis J. 2020;39:894–898.
3. Sevilla-Acosta F, Gutiérrez-Mata A, Yock-Corrales A, et al. Epidemiology, etiology and clinical aspects of childhood acute encephalitis in a tertiary pediatric hospital in Costa Rica. Pediatr Infect Dis J. [Epub ahead of print. 2020]. doi: 10.1097/INF.0000000000002950.
4. Erickson TA, Muscal E, Munoz FM, et al. Infectious and autoimmune causes of encephalitis in children. Pediatrics. 2020;145:e20192543.
5. Domingues RB, Fink MC, Tsanaclis AM, et al. Diagnosis of herpes simplex encephalitis by magnetic resonance imaging and polymerase chain reaction assay of cerebrospinal fluid. J Neurol Sci. 1998;157:148–153.
6. Whitley RJ, Soong SJ, Linneman C Jr, et al. Herpes simplex encephalitis. Clinical Assessment. JAMA. 1982;247:317–320.
7. Graus F, Titulaer MJ, Balu R, et al. A clinical approach to diagnosis of autoimmune encephalitis. Lancet Neurol. 2016;15:391–404.
8. Fowler A, Stödberg T, Eriksson M, et al. Childhood encephalitis in Sweden: etiology, clinical presentation and outcome. Eur J Paediatr Neurol. 2008;12:484–490.
9. Rao S, Elkon B, Flett KB, et al. Long-term outcomes and risk factors associated with acute encephalitis in children. J Pediatric Infect Dis Soc. 2017;6:20–27.
10. Whitley RJ, Kimberlin DW. Herpes simplex encephalitis: children and adolescents. Semin Pediatr Infect Dis. 2005;16:17–23.
11. Falchek SJ. Encephalitis in the pediatric population. Pediatr Rev. 2012;33:122–133.
12. Huang MC, Wang SM, Hsu YW, et al. Long-term cognitive and motor deficits after enterovirus 71 brainstem encephalitis in children. Pediatrics. 2006;118:e1785–e1788.
13. Meligy B, Kadry D, Draz IH, et al. Epidemiological profile of acute viral encephalitis in a sample of Egyptian children. Open Access Maced J Med Sci. 2018;6:423–429.
14. Galanakis E, Tzoufi M, Katragkou A, et al. A prospective multicenter study of childhood encephalitis in Greece. Pediatr Infect Dis J. 2009;28:740–742.
15. Milshtein NY, Paret G, Reif S, et al. Acute childhood encephalitis at 2 tertiary care children’s hospitals in Israel: etiology and clinical characteristics. Pediatr Emerg Care. 2016;32:82–86.
16. Kolski H, Ford-Jones EL, Richardson S, et al. Etiology of acute childhood encephalitis at the hospital for sick children, Toronto, 1994-1995. Clin Infect Dis. 1998;26:398–409.
17. Mailles A, Stahl JPSteering Committee and Investigators Group. Infectious encephalitis in france in 2007: a national prospective study. Clin Infect Dis. 2009;49:1838–1847.
18. Granerod J, Ambrose HE, Davies NW, et al.; UK Health Protection Agency (HPA) Aetiology of Encephalitis Study Group. Causes of encephalitis and differences in their clinical presentations in England: a multicentre, population-based prospective study. Lancet Infect Dis. 2010;10:835–844.
19. de Blauw D, Bruning AHL, Busch CBE, et al.; Dutch Pediatric Encephalitis Study Group. Epidemiology and etiology of severe childhood encephalitis in The Netherlands. Pediatr Infect Dis J. 2020;39:267–272.
20. Whitley RJ. Herpes simplex virus infections of the central nervous system. Continuum (Minneap Minn). 2015;216 Neuroinfectious Disease1704–1713.
21. Klein SK, Hom DL, Anderson MR, et al. Predictive factors of short-term neurologic outcome in children with encephalitis. Pediatr Neurol. 1994;11:308–312.
22. Wang IJ, Lee PI, Huang LM, et al. The correlation between neurological evaluations and neurological outcome in acute encephalitis: a hospital-based study. Eur J Paediatr Neurol. 2007;11:63–69.
23. Lancella L, Esposito S, Galli ML, et al. Acute cerebellitis in children: an elevenyearretrospectivemulticentric study in Italy. Ital J Pediatr. 2017;43:54.
24. Thompson C, Kneen R, Riordan A, et al. Encephalitis in children. Arch Dis Child. 2012;97:150–161.
25. Openshaw H, Cantin EM. Corticosteroids in herpes simplex virus encephalitis. J Neurol Neurosurg Psychiatry. 2005;76:1469.
26. Ramakrishna C, Openshaw H, Cantin EM. The case for immunomodulatory approaches in treating HSV encephalitis. Future Virol. 2013;8:259–272.
27. Gotterer L, Li Y. Maintenance immunosuppression in myasthenia gravis. J Neurol Sci. 2016;369:294–302.
28. Martinez-Torres F, Menon S, Pritsch M, et al.; GACHE Investigators. Protocol for German trial of Acyclovir and corticosteroids in Herpes-simplex-virus-encephalitis (GACHE): a multicenter, multinational, randomized, double-blind, placebo-controlled German, Austrian and Dutch trial [ISRCTN45122933]. BMC Neurol. 2008;8:40.
29. Wang W, Ji M. Efficacy of acyclovir for herpes simplex encephalitis: a protocol for a systematic review of randomized controlled trial. Medicine (Baltimore). 2019;98:e15254.

encephalitis; HSV; cortisone

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.