Acute varicella zoster encephalitis without evidence of primary vasculopathy in a case-series of 20 patients


Corresponding author: A. Mailles, Institut de veille sanitaire (French Institute for Public Health Surveillance), 12 rue du Val d’Osne, 94415 Saint-Maurice Cedex, France


Clin Microbiol Infect


Varicella zoster virus (VZV) is a leading cause of acute viral encephalitis but little is known about its clinical, biological and imaging features. Furthermore, the most favourable treatment regimen has not been determined. We studied a prospective cohort of 20 HIV-negative patients presenting with acute VZV encephalitis caused by primary infection or reactivation. VZV was identified in 16 of 20 cases by PCR detection of the DNA in the cerebrospinal fluid. The four remaining cases occurred during or soon after a VZV rash. The median age of the 17 adults was 76 (19–86) years; the three other patients were children (0.5–5 years). Three patients were immunocompromised. Nine adult patients presented with a rash. Eighteen patients presented with fever and an acute encephalitic syndrome: diffuse brain dysfunction, focal neurological signs, seizures and cranial nerve palsies. Three patients presented with either ventricular or subdural haemorrhage, one with myelitis, and one with asymptomatic stenosis of the middle cerebral artery. The imaging was either normal or revealed non-specific abnormalities such as cortical atrophy but no evidence of stroke. All patients were given acyclovir at various dosages and durations but the case fatality rate remained high (15%) and sequelae were frequently observed either at discharge or at follow-up 3 years later.


Varicella zoster virus (VZV) is a virus of the Alphaherpesvirinae subfamily, responsible for human infections with various clinical presentations. Primary infection occurs most frequently during childhood and presents as varicella; it is followed by long-lasting viral latency in the spinal and cranial ganglia [1,2]. This infection is common, affecting almost all non-vaccinated children. Infection reactivation occurs mainly in immunocompromised or elderly patients. The most frequent presentation in adults is herpes zoster (shingles), with an estimated annual incidence of 1.2–5.2 cases/1000 [2]. Post-herpetic neuralgia is a frequent complication in elderly patients [3].

Neurological complications caused by varicella can involve peripheral or central nervous system. They include cerebellitis, meningitis, myelitis, optic neuritis, polyneuropathy, acute encephalitis and vasculopathy [2,4]. Some neurological presentations can also appear with, precede, or follow herpes zoster: sensory or motor peripheral paralysis, isolated and multiple cranial nerve palsies including ophthalmoplegia and the Ramsay Hunt syndrome, myelitis, encephalitis, vasculopathy and cerebellar ataxia [5]. Neurological symptoms caused by VZV infection can also occur without any cutaneous rash (zoster sine herpete) and the detection of DNA or specific VZV antibodies in the cerebrospinal fluid (CSF) is here the only proof of VZV involvement [5].

Several mechanisms explain the occurrence of neurological signs after VZV infection: parenchymatous encephalitis (with or without demyelination), microglial influx, granulomatous and large vessel vasculopathy inducing strokes (grossly transient cerebral arteriopathy in children and delayed contralateral hemiparesis of zoster ophthalmicus in the elderly), meningitis or necrotizing myelitis [6,7]. Subacute or chronic VZV encephalitis are encountered in immunocompromised patients as the result of small vessel vasculopathy and glial infection inducing leukoencephalitis. Myelitis, large vessel involvement and ventriculitis have also been reported [7,8]. Acute ataxia in children and vasculopathies in adults are the most frequently cited neurological complications of VZV infection [5,9,10]. In contrast, acute encephalitis sensu stricto has rarely been reported [11–14] and even fewer reports have been published describing the clinical features and follow-up of these patients [15–18]. The importance of VZV in the epidemiology of acute encephalitis in the general population has been highlighted in multicentre prospective studies [11–13].

We report the clinical features of patients with VZV encephalitis enrolled in a prospective multicentre study in France, in 2007.


A case-patient with acute encephalitis was a patient 28 days of age or older hospitalized in mainland France in 2007, with (i) an acute onset of illness; (ii) at least one abnormality of CSF (white blood cell count ≥4 cells/mm3 or protein level ≥40 mg/dL); (iii) fever or recent history of fever ≥38°C; and (iv) decreased consciousness, or seizures, or altered mental status, or focal neurological signs. HIV infection was an exclusion criterion. Aetiological investigation of enrolled patients was described previously [11].

A confirmed case of VZV encephalitis was a case-patient with VZV DNA detected in the CSF by PCR. A possible case of VZV encephalitis was a case-patient with encephalitis occurring simultaneously or shortly after varicella or herpes zoster without biological confirmation of VZV infection, and with other possible aetiologies of encephalitis excluded by the investigation. Clinical and biological data were collected using a standardized questionnaire and the results of electroencephalography examinations were recorded. A single expert (TdB) reviewed all available magnetic resonance and computed tomography images to assess brain lesions.

The outcome of VZV patients was assessed 3 years after discharge using the Glasgow outcome scale [19].

Informed written consent was obtained from all patients or from their parents for children aged <18 years. The study was approved by the ethics committee of Grenoble (No. 172003).



Twenty of 253 (8%) case-patients enrolled in the initial study were VZV patients [11]. Sixteen were confirmed cases and four were possible cases (Table 1). Two possible cases were 5-year-old boys with negative VZV CSF PCR on admission and no further CSF testing. They presented with varicella, respectively, 1 day and 1 week before the onset of neurological symptoms. The two other possible cases were adults (76 and 78 years of age) presenting with radicular zoster (one spinal, one ophthalmicus) 3 days and 3 weeks, respectively, before the onset of neurological symptoms.

Table 1.   Characteristics of 20 patients with varicella zoster virus encephalitis, France 2007
Confirmed/PossibleAge (years)SexComorbiditiesVZV rashClinical signsGCSa on admissionWBC count (/mm3)CSF proteins (g/L)CT scanMRIOutcome
  1. GOS, Glasgow outcome scale.

  2. aMaximum value = 10 before 6 months of age, 12 before 1 year of age, 15 afterwards.

  3. bThis patient did not present with any more rash on admission.

Patients with brain vasculopathy
 C0.5MNoneVaricellaFever, vomiting, fontanelle bulging, downward gaze deviation, apathy104000.8D0: Ventricular hydrocephaly, ventricular bleeding
D8: external hydrocephaly
D1: external hydrocephalyRecovery, CSF derivation still in place 3 years later
 C20MNoneNoneFever, headache, incoherent speech, aphasia, facial paralysis, hemiparesis,3500.8 D0: focal stenosis of the left middle cerebral arteryFull recovery
 C79MCardiac insufficiencyShinglesDisorientation, incoherent speech, meningeal syndrome, facial paralysis, severe sepsis14661.1D4 and D22: normal
J27: ventricular bleeding, thalamic and temporal haematoma, brain herniation
D1: left external capsule lacunar sequelaeDied on D34 of hospital stay
 C34MLungs and heart transplant,
Renal insufficiency, Diabetes mellitus
NoneFever, found unconscious at home, seizures, decreased consciousness, brainstem deficits, tetraparesis31005NormalD0: bilateral millimetric high signal T2 lesions of internal capsules, pons and spinal cord at T7–T8
D12 : bilateral subdural hydromas
D16 : normal
D45 :ventricular bleeding and hydrocephaly
Discharged to rehabilitation facility, still paraplegic 3 years later
Patients with parenchymal non vascular lesions
 C75MHistory of renal carcinomaNoneFever, aggressiveness, ideo-motor apraxia, repetitive speech, disorientation, seizures400.8D0: bilateral temporal hypodensitiesFull recovery
 C55MNoneNoneFever, brainstem deficits, tinnitus, ataxia15480.8Hypersignals in brainstem and upper medullaStill suffers from tinnitus, ataxia and aphasia 3 years later
 P5MAdrenal insufficiency
Steroids, Failure to thrive
VaricellabFever, disorientation, agitation, seizures, facial paralysis, hemiparesis, hemianopia79300.2Hypersignal in temporal, parietal, and occipital lobes on T2 weighted MRIBack to previous medical condition
Patients with non-specific lesions on imaging
 C86FCardiac insufficiency
ShinglesFever, more disorientated and incoherent, decreased consciousness, deficit of right leg8301Cortical atrophyDied with acute coma 2 months after discharge
 C77FRespiratory insufficiency, Osteosarcoma, bilateral nephrectomyShinglesFever, agitation, incoherent speech, facial paralysis, ataxia, radiculitis15160.4Cortical atrophyDied of cancer 2 years after discharge
 C85MCardiac insufficiency,
Paraparesis following discal herniation
NoneFever, disorientation, apathy, incoherent speech142001.9Leukoaraiosis, bilateral basal ganglia lacunar sequelaeConsidered cured on discharged, deaf and bedridden 3 years later
 P78MNoneOphthalmic shinglesVomiting, disorientation, incoherent speech, diplopia, aphasia, non-reflexive mydriasis, cerebellous syndrome3301.2Cortical atrophyDischarged with ataxia and disorientation, full recovery 3 years later
Patients with normal imaging results
 P5MNoneVaricellaFever, Cough, decreased consciousness, disorientation, seizures, tetraparesis, gaze deviation1000.7NormalFull recovery
 C26MNoneNoneFever, deafness, ataxia, brainstem deficit, meningeal syndrome3211.6NormalDischarged with partial deafness and facial palsy, total recovery 3 years later
 C20FNoneNoneIncoherent speech, headache, disorientation, meningeal syndrome6001NormalDischarged with memory and attention impairment
 P76FSystemic Lupus erythematous
Hemiplegia (spinal compression) depression
ShinglesDiarrhoea, incoherent speech, pyramidal pathways deficit, decreased consciousness9921NormalBack to previous medical condition (impaired)
 C82MNoneShinglesFever, disorientation, incoherent speech15200.8NormalImproved under treatment then suddenly died at D18 of hospitalization
 C84FNoneShinglesFever, incoherent speech, hyperaesthesia15701NormalGOS = 3
 C63MDepressionNoneFlu-like syndrome, incoherent speech, disorientation, agitation1412403.9J5: normalD7: normal
D18: normal
Full recovery
Patients with no set of images available for review
 C83MRenal, cardiac and lung insufficiencyNoneFever, cough, incoherent speech, disorientation, drowsiness143003Not available Persistent memory disorders and cognitive impairment
 C75MNoneShinglesFever, cough, vomiting, disorientation, incoherent speech, nystagmus, facial palsy, radiculitis, laryngeal paralysis3201.8Not available Died at D23 of hospitalization from aspiration pneumonia

The male to female ratio was 3. Three patients were children (the two 5-year-old boys mentioned above and a 6-month old boy). The distribution of age among adult patients was bimodal with three young adults (19, 20 and 26 years old) and 11 people ≥75 years old (median 76 years old, range 19–86 years). None of the VZV patients had ever been vaccinated against VZV.

Clinical features

Ten (50%) patients (nine adults and one child) had comorbidities, three of which could be considered immunosuppressive conditions: osteosarcoma, recent heart–lung transplant and systemic lupus erythematosus with long-term corticosteroid treatment (Table 1).

Eleven VZV patients (55%) had a rash before hospitalization: zoster in eight adults and varicella in three children, of whom two (a child and an adult) no longer had cutaneous signs on admission.

The symptoms of the VZV patients on admission are presented by frequency in Table 2. Six patients (30%) required admission to intensive care units, four were mechanically ventilated. The most frequent neurological symptoms were disorientation and confusion (70%), meningeal signs (60%), focal neurological signs (55%) and apathy (50%). Cranial nerve palsy was present in eight patients (40%), namely central facial nerve paralysis in seven cases and oculomotor nerve paralysis in one case.

Table 2.   Clinical features of 20 patients presenting with varicella zoster virus encephalitis at day 0 and day 5
 Day 0Day 5
All cases (n = 20)Confirmed cases (n = 16)Possible cases (n = 4)All cases (n = 20)Confirmed cases (n = 16)Possible cases (n = 4)
  1. NA, not available.

  2. aTwo patients were afebrile while taking antipyretic treatment but reported fever before hospitalization.

  3. bOne patient was afebrile while taking antipyretic treatment but reported fever before hospitalization.

Fever18 (90%)a15 (94%)b3 (75%)bNANANA
Mechanical ventilation4 (20%)1 (6%)3 (75%)3 (15%)1 (6%)2 (50%)
Severe sepsis1 (5%)1 (6%)0000
Meningism12 (60%)9 (56%)3 (75%)3 (15%)3 (19%)0
Decreased level of consciousness5 (25%)3 (19%)2 (50%)2 (10%)1 (6%)1 (25%)
Disorientation14 (70%)11 (69%)3 (75%)5 (25%)5 (31%)0
Confusion14 (70%)12 (75%)2 (50%)2 (10%)2 (13%)0
Seizures4 (20%)2 (13%)2 (50%)000
Focal neurological signs11 (55%)7 (44%)4 (100%)9 (45%)7 (44%)2 (50%)
Speech disorders6 (30%)5 (31%)1 (25%)2 (10%)1 (6%)1 (25%)
Cranial nerve palsy8 (40%)6 (38%)2 (50%)8 (40%)6 (38%)2 (50%)
Cerebellous syndrome3 (15%)2 (13%)1 (25%)3 (15%)2 (13%) 
Sensory disorders1 (5%)1 (6%)0NANANA
Myelitis0001 (5%)1 (6%)1 (25%)
Aggressiveness1 (5%)1 (7%)0000
Apathy10 (50%)7 (44%)3 (75%)3 (15%)3 (19%)0
Agitation6 (30%)4 (27%)2 (50%)1 (5%)1 (6%)0
Movement disorders000000

Lumbar puncture was performed before day 2 of hospitalization in 80% of cases and before day 4 in 95%. The median CSF white blood cell count was 150 cells/mm3 (range 0–1240). Lymphocytes were predominant in CSF (median rate: 81.5% of white blood cells) in all but two VZV patients. The median protein level in CSF was 0.99 g/L (0.22–5 g/L). The glucose CSF/serum ratio was >0.40 in 16 patients (80%).

Thirteen of 14 (92%) encephalography recordings showed abnormalities and all such abnormalities were consistent with diffuse brain lesions. Focal temporal slow waves were identified in four cases, and subclinical seizures in two cases.


All patients underwent computed tomography scan (n = 10), magnetic resonance imaging (n = 2), or both (n = 8) on admission. Eighteen sets of images were available for reviewing (14 confirmed cases and four possible cases). Four (20%) patients demonstrated vascular lesions on computed tomography of magnetic resonance imaging:

  •  stenosis of the left middle cerebral artery M1 segment was observed using magnetic resonance angiography in a 20-year-old patient without stroke (Fig. 1);
  •  a 6-month-old boy presented with diffuse bleeding and dilatation of the ventricles on admission and still had a dilatation of the subarachnoid area with normal ventricular size on follow-up imaging (Fig. 2);
  •  thoracic myelitis was observed on admission in a 34-year-old man with paraplegia, who developed at day 45 a bilateral subdural haematoma and massive ventricular bleeding (Fig. 3);
  •  a 79-year-old patient with cortical atrophy on early images later demonstrated a haematoma in the thalamic area with fatal ventricular bleeding at day 23.
Figure 1.

 Magnetic resonance image of a 20-year-old man. Narrowing of the M1 segment of the middle cerebral artery.

Figure 2.

 Magnetic resonance image of a 6-month-old boy. (a, b) Axial T2 gradient echo sequence: blood in the ventricles and internal and external hydrocephaly. (c) Axial SET1 sequence after gadolinium infusion: no abnormal enhancement. (d) FLAIR sequence: no parenchymal abnormality.

Figure 3.

 Magnetic resonance image of a 35-year-old man. (a–c) Cranial magnetic resonance image, FLAIR sequence showing punctiform hyperintensities subcortical (right frontal and left temporal) and in the pons. (d) Sagittal dorsal spinal cord, T2 sequence: hypersignal at the T7–T8 level.

Three (15%) other patients had non-vascular lesions: brainstem lesions suggesting rhombencephalitis were observed in two patients; and right-side temporal, parietal and occipital hypersignals were observed with magnetic resonance imaging in a 5-year-old boy (possible case).

Four patients (20%), including one possible case, had non-specific abnormalities (cortical atrophy) and the remaining seven (35%) patients, including two possible cases, had normal imaging results. Ischaemic stroke was ruled out by imaging in all patients.


Five of the eight adult patients with zoster rash received acyclovir before the onset of neurological symptoms. Acyclovir was administered intravenously to all 20 patients as encephalitis treatment for 3 weeks (n = 8; 40%), 2 weeks (n = 5; 25%) or 1 week (n = 7; 35%). Dosage was reported for 13 of 20 patients: five received 10 mg/kg three times a day, six received 15 mg/kg three times a day and two children received 20 mg/kg three times a day.


Three male VZV patients (15%) died during hospitalization: all three were elderly (75, 79 and 82 years old). The mean duration of hospital stay was 23 days (range 6–80). On discharge, 10 patients (50%) returned home, six patients (30%) were transferred to a convalescence facility, and one moved to a nursing home.

Nine patients (45%) were discharged with persistent neurological signs: cognitive impairment (n = 3) and sensory-motor deficits (n = 6). Three years after discharge, 16 of the 17 surviving patients could be evaluated and one was lost to follow-up. Among the 16 patients, two had died after being discharged (with one death being related to the encephalitis), seven (41%) had moderate to severe sequelae (Glasgow outcome scale 3 or 4), and the outcome was favourable for seven (41%) with a Glasgow outcome scale of 5.


Varicella zoster virus has been confirmed as a major cause of encephalitis in France both in the national hospital database and a prospective cohort study [11,20], as well as in other countries [12,13].

The data described here highlight the main clinical features of VZV encephalitis. VZV encephalitis is a disease of the young and elderly immunocompetent patients, as well as a disease of the immunocompromised of all ages, as illustrated by four young adults enrolled in our study.

Varicella zoster virus encephalitis has been described during the course of or after varicella or zoster with various time frames. However, the rash was absent in nine of 17 adult patients enrolled in our study with clinical encephalitis and a positive VZV PCR in CSF. Conversely, no virological confirmation could be obtained in four patients whose encephalitis occurred during or shortly after varicella or zoster. In these four patients, the clinical signs, imaging and outcome strongly suggested a causative link between VZV and the encephalitis, and the negative result of the PCR might be related to the early CSF sampling, or to the administration of acyclovir before CSF sampling [17,21]. The possible lack of sensitivity of the PCR has been demonstrated for herpes simplex virus during the first days of neurological signs but not formally for VZV. In these four patients, it may have been interesting to have VZV antibodies measured in a convalescent sample of CSF to try to confirm the diagnosis [21].

Few studies have been published about patient with VZV encephalitis (Table 3). In a paediatric study, the case-fatality rate among encephalitis patients was 34%, which is higher than in our patients but this study was published before molecular diagnosis and acyclovir were available [15]. In a 12-patient cohort reported in 1983, ten patients had their neurological onset a mean of 9 days after a zoster rash and two patients reported no rash, which is comparable to our study [16]. Their case fatality rate was 25%, but there again the study was carried out before acyclovir was widely available. In a more recent study, CSF samples from encephalitis patients were screened for VZV and seven patients were diagnosed with VZV encephalitis during a 5-year period. Their demographic characteristics (especially the bimodal distribution of age), and their clinical signs were similar to those observed in our prospective study [18].

Table 3.   Clinical, biological and imaging features of varicella zoster encephalitis patients included in the main published series
Year of diagnosisNumber of casesAge of casesImmune status of casesDiagnosis/case definitionRashDelay between rash and neurological onsetClinical patternsImaging resultsEncephalitis/
1956–196753Mean 5 years (range 8 months to 8.5 years)NDNeurological signs + varicellaRash is part of the case definitionMean 4 days (range 1–20)23 with encephalitis, 29 with cerebellitis, one with myelitisNo imaging2/6 cases with necropsy showed signs of perivascular inflammationEncephalitis: eight deaths, two discharged with permament sequelae.
Cerebellitis: seven full recoveries, no death.
Myelitis: full recovery
Before brain imaging, PCR and acyclovir15
1971–197812Median 71 years (range 23–80)Seven immunosuppressedEncephalitis or cerebellitis and herpes zosterHerpes zoster is part of the case definitionMean: 9 days (max 28 days)Hallucinations 10/12
Stupor 5/12
Confusion, headache, cranial nerve palsy
1/5 with abnormal results on CT scan (ventricular enlargement) Three deaths, one case with permanent paresisBefore MRI, PCR and acyclovir16
19752(1) 19 years old
(2) 55 years old
Both immunodepressed following cancerClinicalHerpes zoster in bothND(1) Cognitive decline, leg weakness, poor memory
(2) Seizures, hemiparesis, poor memory
Both cases: parenchymal mass effect, evolution toward haemorrhageBoth cases: encephalitis and secondary vascularitis(1) Death after 6 weeks
(2) death after 6 months
Before acyclovir and MRI and PCR22
1978–19822(1) 42 years old
(2) 62 years old
Both immunocompromised (leukaemia and renal transplant)Neurological signs + herpes zosterHerpes zoster is part of the case definition(1) 10 weeks
(2) At onset
(1) Ataxia, poor memory, hemiparesis, neuralgia
(2) Hemiparesis, cranial nerves alsy, hemianopsie
Normal in one patient, mass effect for the otherNo evidence of vascularitisFavourable outcome in both casesBefore acyclovir, MRI and PCR6
1985–1995925–73 years oldImmunocompetentSerology on CSF or serum, viral culture of skin lesions5/9 cases, 4–8 days before neurological onset4–8 daysFever 3/9
Headache 5/7
Speech disorders 5/7
Seizures 4/7
Motor deficit 3/7
1 ‘infarct-like lesion’, one non-specific lesion, other normal1 ‘infarct-like lesion’On discharge: Poor memory 7/9, disinhibited behaviour 6/9, poor concentration 3/9 23
1990–200424Mean 2 years and 9 months (range 2 months to 6 years)All immunocompetentBrain infarction + varicella within 12 monthsVaricella is part of the case definitionMean 4 months (range 1 week to 1 year)Hemiparesis 23/24
Chorea 1/24
Facial weakness 1/24
Seizure 2/24
Caudate and lentiform nuclei 21/24
Internal capsule 13/24
White matter infarction 19/24
Cortical infarction 8/24
Infarction is part of the case definitionND 24
1992–19964NDAll patients with AIDSDNA detection + clinical signs + consistent neuroimaging2/4 with herpes zosterNDOne encephalitis, one meningoencephalitis, two focal encephalitisOne with haemorrhagic lesion, one with ischaemic lesion, one with cortical atrophy, one with brainstem lesionsTwo with vascularitis, two with encephalitisTwo deaths and two complete recoveries 25
1995–199692All agesNDClinical signs + serology or DNA detectionHerpes zoster in 37, varicella in 30, none in 25Max 4 weeks for varicellaNDNDNDNDNational study on CNS viral infections17
1995–200628Median 72 years (range 3–86)NDSerology and RT-PCRRash in 19/28NDND6/28 with haemorrhage or infarctionSix with cerebrovascular lesionsTwo deaths, three patients with sequelaeVascular lesions are not detailed26
1998–200911Median age 75 years (range 50–85)5/11 immunodepressed (two with AIDS), 9/11 immunosenescentNeurological signs + DNA detection in CSF7/11 (64%)NDAtaxia 4/11
Cranial nerve palsy 4/11
Fever 5/11
Altered mental status 9/11
Vasculopathy 2/11
Parenchymatous lesions 2/11
Normal Imaging or non-specific lesions 7/11
Two cases with evidence of vasculopathyND 21
2000–20012(1) 4 years old
(2) 16 months old
Both immunocompetentSerology + intrathecal antibody synthesis + DNA detection(1) Varicella
(2) None
1 week(1) Focal seizure, confusion
(2) Fever, irritability, confusion
(1) Subcortical lesion in temporal lobe, no stroke
(2) Cortical and subcortical lesion of parietal lobe, White matter lesions
No stroke, no evidence of vasculopathy(1) Full recovery
(2) Permanent hemianopsia
2004–20097Mean 47 years (range 14–79), bimodal distribution2/7 immunodepressedClinical signs + DNA detection in CSFNDNDMost frequent: fever, headache, neck stiffness, photophobiaNormal imaging in all patientsNo evidence of stroke or vasculopathyOne death (terminal renal failure) 18
ND9Median 72 years (range 52–94)All immunocompetentNeurological signs + rashHerpes zoster is part of the case definitionNDAtaxia n = 5, cranial nerves impairment n = 2CT scan in 5/9, normal for all 5No evidence of vascularitisNo death, sequelae not describedApparently before PCR and MRI but acyclovir available28
ND30Range 1–88 years old11/30 immunocompromisedNeurological signs and Imaging or CSF consistent with vasculopathy and DNA detection or serology19/30 (63%)NDNDLarge-vessel vasculopathy 4/30 (13%)
Small-vessel vasculopathy 11/30 (37%)
Mixed vasculopathy 15/30 (50%)
Vasculopathy is part of the case definitionFavourable outcome for four cases with large-vessel vasculopathy
Poor outcome for three cases with small-vessel and three cases with mixed vasculopathies
23 previously published cases + seven newly published cases29

Twenty-six patients with positive CSF VZV DNA were referred to the California Encephalitis Project, 11 suffering encephalitis [30]. Some features were similar to those of our patients (median age 75 years, 73% of imaging was normal or non-specific), although a zoster rash was more frequent (64%).

Varicella zoster virus encephalitis is known to be associated with immunosuppression, especially that caused by HIV. Because HIV infection was an exclusion criterion, it is likely that we missed some patients with VZV encephalitis and HIV infection. However, the case reports published about VZV encephalitis in patients with AIDS show that their clinical presentation does not differ from that of immunocompromised patients following cancer or long-term immunosuppressive treatment.

No study or clinical trial has yet addressed the best therapeutic regimen for VZV encephalitis. The prognosis of herpes simplex virus encephalitis has been dramatically improved by the use of acyclovir and this treatment is recommended in the guidelines for VZV encephalitis [2]. The recommended regimen is acyclovir at the same dosage as herpes simplex virus encephalitis treatment despite the lack of paediatric data [14,31,32]. Unlike herpes simplex virus, few studies have been published on VZV encephalitis [33]. In our study, 3 years after discharge, half of the survivors still presented with moderate to severe sequelae. Some authors have suggested that the combination of acyclovir and foscarnet might improve the prognosis and also prevent a possible antiviral failure caused by an acyclovir-resistant VZV strain, but this hypothesis has yet to be confirmed by clinical trial [34].

Questions and controversies remain on the physiological mechanisms of the neurological complications of VZV infection. These complications are different during primary infection (varicella) and after viral reactivation (zoster). The most frequent varicella complications are cerebellitis and arterial ischaemic strokes [15,35,36]. Usually, VZV cerebellitis or ataxia has a favourable outcome in children and CSF analysis is generally not performed in this context. Besides ataxia, arterial ischaemic stroke occurs in up to 1/15 000 cases [36]. This syndrome, now referred to as post-varicella arteriopathy, is defined by focal stenosis of the basal central arteries in children with a history of varicella within the 12 months before the onset of neurological signs.

Varicella acute encephalopathy with fatal outcome has also been reported following varicella in children with liver and brain oedema and fatty macrophage infiltration as the result of Reye’s syndrome, although this does not actually correspond to a direct invasion of the central nervous system [15].

The physiopathology of VZV encephalitis after reactivation of the virus is still not clear. The presentations of herpes zoster encephalitis can be divided into three main categories: demyelinating disease, vasculopathy and acute infectious encephalitis of undetermined pathophysiology, as reported in this case series. The multifocal subacute demyelinating disease is mostly encountered in immunosuppressed patients, especially those with AIDS (which was an exclusion criterion in our study). The direct infection of glial cells and endothelial vascular cells has been demonstrated in these cases [7]. The VZV-induced vasculopathy occurs in immunocompetent elderly people or in AIDS patients (for instance, zoster ophthalmicus contralateral syndrome) [8,29]. Although some authors claim that most, if not all, cases of herpes zoster ‘encephalitis’ are the result of viral vasculopathy [5,9], imaging failed to demonstrate either vascular lesions or demyelinating areas in 16 of 20 patients enrolled in our study, which is in favour of direct viral parenchymatous encephalitis.

However, one patient in our series presented with both myelitis and vasculopathy. He was under immunosuppressive treatment following heart and lung transplant, and suffered renal deficiency and diabetes. Myelitis has been described as a non-vascular clinical presentation, and it has been hypothesized that long-term steroids might be a risk factor for VZV myelitis [1]. The clinical features and imaging results in this patient suggest the simultaneous occurrence of two different clinical presentations: first the myelitis, and second the vasculopathy. We can hypothesize that the immunosuppressive condition of this patient favoured the persistence of a high viral load in the brain arteries, leading to a massive vasculopathy both in subdural and ventricular areas.

Unlike herpes simplex virus, the correlation between the presence of VZV in CSF and brain infection has not been demonstrated. However, it has been demonstrated that the VZV viral load in the CSF was correlated to the severity of central nervous system symptoms in patients presenting with encephalitis [37]. Moreover, the acyclovir-induced clearance of VZV DNA from the CSF was associated with clinical improvement in four cases of encephalitis [26]. These findings are in favour of a direct viral encephalitis besides vascularitis and demyelinating encephalitis. However, some crossover between acute VZV encephalitis and VZV-induced vasculopathy cannot be excluded, and both mechanisms might exist in the same patients following reactivation of the virus. This is suggested by the observation of a patient enrolled in our study with a large vessel vasculopathy on angiography and normal brain imaging excluding both a stroke and a demyelinating process.


Varicella zoster virus encephalitis is the second leading cause of acute infectious encephalitis in France, accounting for nearly 10% of all cases. According to our results, this diagnosis should be considered in any case of central nervous system acute febrile disease with lymphocytic aseptic meningitis, especially in elderly and immunocompromised patients. The absence of any rash, as in more than half of the cases in our study, should not be considered as evidence excluding the diagnosis of VZV encephalitis. Despite the low level of evidence, acyclovir treatment should be prescribed to these patients, because of its antiviral effectiveness against VZV. The case fatality rate remains high and sequelae are frequent. Controversies remain about its pathophysiology and further research should be undertaken to determine the optimal therapeutic regimen.


TdB, AM, SB, PM and JPS contributed equally in the redaction of the manuscript. AM was responsible for the analysis of data. The members of the steering committee contributed to the critical reviewing of the manuscript. The members of the investigators group contributed to the investigation and management of the patients, and to the collection and interpretation of data.

Transparency Declaration

The Institut de veille sanitaire (French institute for public health surveillance), Saint Maurice, France promoted the study and funded the salary of two data collection personnel. Glaxo SmithKline, Roche and Biomérieux funded the setting up and maintenance of a biobank of samples taken from the patients. All authors have no conflict of interest to disclose.


Steering committee

Cecile Bébéar (Bordeaux), Cecile Brouard (Saint-Maurice), Thomas De Broucker (Saint-Denis), Eric Cua (Nice), Henri Dabernat (Toulouse), Daniel Floret (Lyon), Benoit Guery (Lille), Marc Lecuit (Paris), Bruno Lina (Lyon), Olivier Lortholary (Paris), Alexandra Mailles (Saint-Maurice), Christian Michelet (Rennes), Patrice Morand (Grenoble), Bruno Pozzetto (Saint-Etienne), Jean-Paul Stahl (Grenoble), Veronique Vaillant (Saint-Maurice), Yazdan Yazdanpanah (Tourcoing), Herve Zeller (Lyon).


Philippe Abboud (Rouen), Chakib Alloui (Paris), Christine Archimbaud (Clermont-Ferrand), Bruno Barroso (Pau), Louis Bernard (Garches), Pascal Beuret (Roanne), Geneviève Billaud (Lyon), Thierry Blanc (Rouen), Michèle Bonnard-Gougeon (Clermont-Ferrand), David Boutolleau (Paris), Cédric Bretonnière (Nantes), Céline Bressollette-Bodin (Nantes), Fabrice Bruneel (Versailles), Marielle Buisson (Dijon), Anne Caramella (Nice), Bernard Castan (Auch), Isabelle Cattaneo (Bry sur Marne), Charles Cazanave (Bordeaux), Stéphane Chabrier (Saint-Etienne), Marie-Laure Chadenat (Versailles), Martine Chambon (Clermont-Ferrand), Pascal Chavanet (Dijon), Mondher Chouchane (Dijon), Pierre Clavelou (Clermont-Ferrand), Pierre Courant (Avignon), Eric Cua (Nice), Fabienne de Brabant (Montélimar), Arnaud De La Blanchardière (Caen), Geoffroy De La Gastine (Caen), Henri De Montclos (Bourg-en-Bresse), Eric Denes (Limoges), Philippe Desprez (Strasbourg), Anny Dewilde (Lille), Aurelien Dinh (Garches), François Durand (Saint-Etienne), Guillaume Emeriaud (Grenoble), Olivier Epaulard (Grenoble), Giovanni Favaretto (Avranche), Anna Ferrier (Clermont-Ferrand), Vincent Foulongne (Montpellier), François Fourrier (Lille), Véronique Gaday (Pontoise), Jacques Gaillat (Annecy), Serge Gallet (Montluçon), Nicole Gazuy (Clermont-Ferrand), Stéphanie Gouarin (Caen), Pascale Goubin (Caen), Alain Goudeau (Tours), Joel Gozlan (Paris), Philippe Granier (Bourg-en-Bresse), Isabelle Gueit (Rouen), Amélie Guihot (Paris), Christine Guillermet (Besançon), Christelle Guillet-Caruba (Paris), Yves Guimard (Bourges), Yves Hansmann (Strasbourg), Cécile Henquell (Clermont-Ferrand), Jean-Louis Herrmann (Garches), Jérome Honnorat (Lyon), Nadhira Houhou (Paris), Benoit Jaulhac (Strasbourg), Olivier Join-Lambert (Paris), Manoelle Kossorotoff (Paris), Emmanuelle Laudrault (Montélimar), Frédéric Laurent (Lyon), Jean-Jacques Laurichesse (Paris), Sylvain Lavoue (Rennes), Leila Lazaro (Bayonne), Stephane Legriel (Versailles), Olivier Lesens (Clermont-Ferrand), Gérard Level (Verdun), Muriel Mace (Orléans), Bénédicte Maisonneuve (Montluçon), Alain Makinson (Montpellier), Hélène Marchandin (Montpellier), Laurent Martinez-Almoyna (Saint-Denis), Patrick Marthelet (Montélimar), Martin Martinot (Colmar), Bruno Massenavette (Lyon), Laurence Maulin (Aix-en-Provence), Benoit Misset (Paris), Catherine Neuwirth (Dijon), Florence Nicot (Toulouse), Jérome Pacanowski (Paris), Jean-Bernard Palcoux (Clermont-Ferrand), Patricia Pavese (Grenoble), Thomas Perpoint (Lyon), Martine Pestel–Caron (Rouen), Robin Pouyau (Lyon), Thierry Prazuck (Orléans), Virginie Prendki (Paris), Christophe Rapp (Saint-Mandé), Christel Regagnon (Clermont-Ferrand), Matthieu Rigal (Auch), Nathalie Roch (Grenoble), Olivier Rogeaux (Chambéry), Sylvie Rogez (Limoges), Charles Santre (Annecy), Anne Signori-Schmuck (Grenoble), Fabrice Simon (Marseille), Abdelilah Taimi (Roanne), Jérome Tayoro (Le Mans), Daniel Terral (Clermont-Ferrand), Audrey Therby (Versailles), Francis Vuillemet (Colmar).