lable at ScienceDirect

Pediatric Neurology 49 (2013) 191e194

Contents lists avai

Pediatric Neurology

journal homepage: www.elsevier .com/locate/pnu

Original Article

Viral-Induced Intracranial Hypertension Mimicking Pseudotumor Cerebri

Sarit Ravid MDa,*, Yael Shachor-Meyouhas MDb, Eli Shahar MDa, Zipi Kra-Oz PhD c,Imad Kassis MDb

a Pediatric Neurology Unit & Epilepsy Service, Meyer Children’s Hospital, Rambam Health Care Campus, Haifa, Israelb Pediatric Infectious Disease Unit, Meyer Children’s Hospital, Rambam Health Care Campus, Haifa, IsraelcMedical Virology Laboratory, Meyer Children’s Hospital, Rambam Health Care Campus, Haifa, Israel

article information

Article history:Received 7 January 2013Accepted 10 March 2013

a

Bno

* Communications should be addressed tNeurology Unit; Meyer Children’s HospitalCampus; Haifa 31096, Israel.

E-mail address: s_ravid@rambam.health.gov.il

0887-8994/$ - see front matter � 2013 Elsevier Inc. Ahttp://dx.doi.org/10.1016/j.pediatrneurol.2013.03.007

bstract

ACKGROUND: Pseudotumor cerebri or idiopathic intracranial hypertension is characterized byormal spinal fluid composition and increased intracranial pressure in the absence of a space-ccupying lesion. METHODS: This study describes a subgroup of 10 patients with the same

typical presenting symptoms (headache, vomiting, and papilledema) but without nuchalrigidity, meningeal signs, or change in mental status. Patients had normal neuroimagingstudies and intracranial hypertension but also pleocytosis in the cerebrospinal fluid, sug-gesting central nervous system infection. From the results it can be hypothesized that thosechildren represent a unique subgroup of viral-induced intracranial hypertension whencomparing their risk factors, clinical course, treatment, and outcomewith 58patientswhohadidiopathic intracranial hypertension. RESULTS: All patients with viral-induced intracranialhypertension presented with papilledema but none had reduced visual acuity or abnormalvisual fields, compared with 20.7% of patients who had idiopathic intracranial hypertension.They also responded better to treatment with acetazolamide, needed a shorter duration oftreatment (7.7 � 2.6 months vs 12.2 � 6.3 months, P ¼ 0.03), and had no recurrences.CONCLUSIONS: The results suggest that children who fulfill the typical presenting signs andsymptoms and all diagnostic criteria for pseudotumor cerebri other than the normal cere-brospinal fluid component may represent a unique subgroup of viral-induced intracranialhypertension and should be managed accordingly. The overall prognosis is excellent.

� 2013 Elsevier Inc. All rights reserved.

Introduction

Pseudotumor cerebri or idiopathic intracranial hyper-tension is defined as elevated intracranial pressure withnormal cerebrospinal fluid (CSF) composition and exclusionof structural abnormalities and systemic disorders [1]. Thissyndrome, initially called serous meningitis, was first de-scribed byQuincke [2] in 1897 inpatientswho had increasedintracranial pressure without a brain tumor. The diagnosticcharacteristics of this syndrome were first described byDandy [3] in 1937 and were later modified into diagnosticcriteria by Smith [4] in 1985. Over the past several years,

o: Dr. Ravid; Pediatric; Rambam Health Care

ll rights reserved.

advances in research and diagnostic technology haveprompted physicians to suggest expansion of the originalcriteria, such as diagnostic opening pressure, or a statementto the effect that no other cause of intracranial hypertensioncould be identified [5]. Although normal CSF composition isone of the essential criteria, occasional reports have beenpublished on patients who have been diagnosed withpseudotumor cerebri even when the CSF showed somepleocytosis [6-8]. These patients fulfilled all the othercriteria for pseudotumor cerebri and, in the absence of anyother cause of intracranial hypertension, were diagnosedand treated as such. Even in the original article by Dandy [3],two of his patients had an abnormal cell count in the CSF.These patients had the same clinical presentation and thesame outcome as the other 20 patients.

The main objective of this study was to compare pre-senting features, risk factors, treatment, and prognosis

S. Ravid et al. / Pediatric Neurology 49 (2013) 191e194192

among patients with intracranial hypertension, with andwithout normal CSF composition, and to determine ifnormal CSF composition is indeed an indispensable criteria.

Methods

This is a retrospective study of pediatric patients aged 3-18 years whowere diagnosed with intracranial hypertension at Meyer Children’sHospital, a tertiary pediatric center in the north of Israel, from 2001 to2010. The control group included patients who were diagnosedaccording to the modified Dandy criteria [4]: (1) symptoms and signs ofincreased intracranial pressure; (2) normal neurological examinationwith the exception of papilledema, visual loss, or sixth nerve palsy; (3)absence of a mass lesion or hydrocephalus; and (4) CSF opening pressureby lumbar puncture higher than 20 cm H2O for children <8 years of ageand 25 cm H2O in children �8 years, with normal fluid parameters(n ¼ 58). Patients who fulfilled all the Dandy criteria apart from normalcomposition of cerebrospinal fluid (n ¼ 10) were placed in the studygroup. Patients with the diagnosis of sinus vein thrombosis or othercauses that obstructed venous outflow were excluded from the study.

Patient records were reviewed for presenting signs and symptoms,brain imaging, CSF opening pressure and composition, treatment ofincreased intracranial pressure, and short-term prognosis. Associatedrisk factors, such as obesity (body mass index >95%), drugs, andhormones, were also evaluated. Outcome measures included duration oftreatment and recurrence rate. These parameters were analyzed andcompared between the two groups.

Statistical analysis

Data were summarized as proportions or means and standard devi-ations. Chi-square analysis was used to test for qualitative variables, andStudent’s t test was used for quantitative variables. For all comparisonsand analyses a P value of <0.05 was used as the cutoff point of statisticalsignificance.

The studywas approved by the institutional review board committee.

Results

The study population consisted of 68 children: 30 (44.1%)boys and 38 (55.9%) girls. Based on the CSF component, twogroups were identified: group 1 (n ¼ 58), which includedchildren with intracranial hypertension and normal CSFresults, and group 2 (n ¼ 10), which included children withintracranial hypertension and CSF pleocytosis (mean, 78.6white blood cells/mm3, range, 18-276). Blood and CSFcultures were sterile in both groups. Two patients, one fromgroup 1 and one from group 2 (Patient 9), were previouslyreported as having manifestation of varicella zoster virus

Table 1. Clinical characteristics of patients with intracranial hypertension and abno

Patient No. Age (yr) Sex Presenting Symptoms CSF Opening Pr

1 8 F HA, vomiting 292 12 F HA, vomiting, fever 363 4 M HA, dizziness 344 14 M HA, dizziness 425 11 M HA, diplopia, blurry vision 396 10 M HA, vomiting, fever 407 4 F HA, vomiting 328 7 M HA, vomiting, fever 309* 9 M HA, vomiting, photophobia 38

10 6 F HA, vomiting, fever 36

Abbreviations:CSF ¼ Cerebrospinal fluidHA ¼ HeadacheWBC ¼ White blood cells

* Patient was previously reported as manifestation of varicella zoster virus reactivatio

reactivation [9]. The clinical characteristics of patients withintracranial hypertension and abnormal CSF compositionare summarized in Table 1.

The mean age of symptoms’ onset in group 1 was 9.3 �2.7 years (median, 10.5 years, range, 3-18), compared with8.7 � 3.7 years (median, 9.5 years, range, 4-16) in group 2(NS). In group 1, 34 children were girls (58.6%), comparedwith 6 (60%) in group 2 (NS). Duration of symptoms beforepresentation ranged between 3 and 13 weeks (median, 5.6weeks) in group 1, compared with 2 to 5 weeks (median, 3.2weeks) in group 2.

The most common presenting symptom reported forboth groups was headaches (Table 2), followed by vomiting.Visual obscurations and blurry visionweremore common inchildren with normal CSF component (17.2% vs 10%in groups 1 and 2, respectively), but without statisticalsignificance. Optic disc edema was present in all of thesepatients, while only approximately 10% of the patients inboth groups experienced sixth nerve palsy. Visual fields andacuity were normal in all children from group 2. Twelve(20.7%) children in group 1 had abnormal visual fields andthree (5.2%) had decreased visual acuity that was related tothe increased pressure.

All patients had normal brain imaging, either computedtomography or magnetic resonance imaging. There was nosignificant difference in lumbar puncture opening pressurebetween the two groups (35.6 � 4.3 vs 34.7 � 5.2 cm H2O).

Infection was the most common risk factor in group 2. Aconfirmed etiologic agent was identified in four patients,two had enterovirus, one had Epstein-Bar virus, and onehad varicella zoster virus. The diagnosis was made bystandard polymerase chain reaction method. Two patientsin group 1 had evidence of varicella infection with normalCSF values. Other risk factors for pseudotumor cerebri, suchas obesity, drugs (e.g., minocycline), and growth hormoneand iron deficiency anemia, were significantly morecommon in group 1 than in group 2 children (67.2% vs 10%;P < 0.017). Obesity was the most common risk factor ingroup 1, where 30 of 58 children were found to be obese,followed by medications in four of 58 children and irondeficiency anemia in three of 58 children.

All children were initially treated with acetazolamide,which was effective for the majority of patients in bothgroups (80.9%). Six patients from group 2 were treatedwith intravenous ceftriaxone pending culture results.

rmal cerebrospinal fluid composition

essure (cm H2O) CSF Cells (WBC/mm3) Treatment

18 Acetazolamide40 Acetazolamide, ceftriaxone26 Acetazolamide48 Acetazolamide

108 Acetazolamide, ceftriaxone60 Acetazolamide, ceftriaxone64 Acetazolamide80 Acetazolamide, ceftriaxone

276 Acetazolamide, ceftriaxone36 Acetazolamide, ceftriaxone

n [9].

Table 2. Comparison of symptoms and signs between patients with intracranial hypertension with normal and abnormal cerebrospinal fluid composition

Symptom/Sign Intracranial Hypertension Normal CSF N (%) Intracranial Hypertension Abnormal CSF N (%) P

Headache 54 (98.3) 10 (100) NSNausea/vomiting 42 (72.4) 7 (70) NSBlurred vision 10 (17.2) 1 (10) NSDiplopia 6 (10.3) 1 (10) NSPhotophobia 4 (7) 1 (10) NSPapilledema 58 (100) 10 (100) NSAbnormal visual fields 12 (20.7) 0 (0) NSFever 0 (0) 4 (40) <0.001Dizziness 10 (17.2) 2 (20) NSCSF opening pressure (cm H2O) 35.6 � 4.3 34.7 � 5.2 NS

Abbreviation:CSF ¼ Cerebrospinal fluid

S. Ravid et al. / Pediatric Neurology 49 (2013) 191e194 193

Eleven (19%) patients from group 1 but none from group 2received an additional short-term treatment with intra-venous steroids. Lumboperitoneal shunts were placed inonly two patients, both from group 1, due to refractorysymptoms and decreased visual acuity. The duration ofhospitalization was similar in both groups (5.7 � 0.3 daysvs 6.2 � 0.5 days in groups 1 and 2, respectively [NS]).Duration of treatment was significantly shorter in group 2(7.7 � 2.6 months vs 12.2 � 6.3 months in group 1;P ¼ 0.03). End of treatment was determined as resolutionof symptoms and no evidence of optic disc edema of atleast 1 month. The mean follow-up period was 19.7 � 3.9months in group 1 and 14.2 � 2.8 months in group 2. Fourpatients in group 1 had single recurrent episodes, and onepatient had two episodes. No patient in group 2 had anyrecurrence.

Discussion

Idiopathic intracranial hypertension is characterizedby increased intracranial pressure and normal CSFcomposition, in the absence of a space-occupying lesionor hydrocephalus [1]. Presenting symptoms are headache,vomiting, visual obscuration, and diplopia. Bilateral pap-illedema is found in most instances [10,11]. The smallgroup of 10 patients in this study presented with thetypical clinical signs and symptoms of intracranialhypertension, including bilateral optic disc edema. Mostwere afebrile and none had nuchal rigidity, meningealsigns, or change in mental status. These patients hadabnormal CSF examinations but fulfilled all the othermodified Dandy criteria [4]. The main differential diag-nosis on presentation was viral meningoencephalitis, butthe presence of bilateral papilledema, the hallmarkfinding in intracranial hypertension and a very rare signof meningoencephalitis, and the absence of any othertypical signs and symptoms of meningitis or encephalitis,such as somnolence or encephalopathy, led to the diag-nosis of viral-induced intracranial hypertension. In hisoriginal article, Dandy [3] described two of 22 patientswith abnormal cerebrospinal fluid and suggested that aninflammatory process may be considered a possiblemechanism for the pathogenesis of the disorder. Ducloset al. [7] described a 15-year-old girl who developeddecreased visual acuity, bilateral papilledema, sixth nerve

palsy, increased CSF pressure, and lymphocytosis. Theysuggested that the intracranial hypertension wassecondary to viral meningitis.

The association between idiopathic intracranialhypertension and infections has been described [11].Most common is the association with middle ear infec-tions, especially before the use of wide-spectrum antibi-otics. In most instances, this was due to dural sinusthrombosis, mainly of the lateral sinus [12]. The associa-tion between idiopathic intracranial hypertension andviral infections is rare and only isolated cases have beendescribed, including hepatitis E virus [13], hepatitis Avirus [14], and measles [15]. Individuals with idiopathicintracranial hypertension secondary to varicella infectionor as the sole manifestation of varicella reactivation havealso been described [9,16].

The pathogenesis and the relation of idiopathic intra-cranial hypertension to infections are unknown. Animmune-mediated postinfection vasculitis was suggestedas a possible explanation in varicella [17]. CSF pleocytosisoccurred in a few of the individuals with varicella-associated intracranial hypertension [6,9] but not in theother viral infections that have been associated with idio-pathic intracranial hypertension. The fact that the CSF isnormal, even in those patients with proven viral infection,rules out direct infectious injury and suggests a reactiveprocess, either immune mediated or other.

Our data suggest that there is a small subgroup ofpatients with the combination of symptoms of intracranialhypertension, including optic disc edema and CSF pleocy-tosis. The presence of CSF pleocytosis in these patients mayreflect a different pathogenesis of the same clinicalsyndrome and not a different disease. The study resultssuggest that an inflammatory process, either primaryduring acute central nervous system infection or secondaryto an immune-mediated process, may cause dysfunction ofthe absorptive mechanism of the arachnoids’ granulations,ending in decreased central nervous system absorption andintracranial hypertension. This hypothesis might alsoexplain the minority of risk factors, other than infection, inthis paucity of patients and the good response to the samemedical treatment. The fact that none of the patients withabnormal CSF received an additional short-term treatmentwith intravenous steroids, as well as a shorter duration oftreatment and no recurrences, is more supportive of a direct

S. Ravid et al. / Pediatric Neurology 49 (2013) 191e194194

infectious process that is self-limited, and recovery occursin parallel with recovery from the infection.

Conclusions

Children who fulfill the typical presenting signs andsymptoms and all diagnostic criteria for pseudotumor cer-ebri other than a normal CSF component may representa unique group of viral-induced intracranial hypertension.Investigations for various viral etiologies should be con-ducted and treatment should include the same medicationsrecommended for idiopathic intracranial hypertension. Theoutcome appears to be better than it is in the classic form.

References

[1] Ball AK, Clarke CE. Idiopathic intracranial hypertension. LancetNeurol 2006;5:433e42.

[2] Quincke H. Uber Meningitis serosa and verwandte Zustande. DtschZ Nervenheilkd 1897;9:149e68.

[3] Dandy WE. Intracranial pressure without brain tumor: Diagnosisand treatment. Ann Surg 1937;106:492e513.

[4] Smith JL. Whence pseudotumor cerebri. J Clin Neuroophthalmol1985;5:55e6.

[5] Friedman DI, Jacobson MD. Diagnostic criteria for idiopathicintracranial hypertension. Neurology 2002;59:1492e5.

[6] Millichap JJ, Freeman JL. Zoster-associated intracranial hyperten-sion: A case report. Pediatr Neurol 2005;32:211e2.

[7] Duclos C, Bonnin N, Merline E, et al. Benign but not harmlessintracranial hypertension: A case report. Arch Pediatr 2010;17:1677e81.

[8] Kan L, Sood SK, Maytal J. Pseudotumor cerebri in Lyme disease: Acase report and literature review. Pediatr Neurol 1998;18:439e41.

[9] Ravid S, Shachor-Meyouhas Y, Shahar E, Kra-Oz Z, Kassis I.Reactivation of varicella presenting as pseudotumor cerebri:Three cases and a review of the literature. Pediatr Neurol 2012;46:124e6.

[10] Hacifazlioglu Eldes N, Yilmaz Y. Pseudotumor cerebri in children:Etiological, clinical features and treatment modalities. Eur J Pae-diatr Neurol 2012;16:349e55.

[11] Standridge SM. Idiopathic intracranial hypertension in chil-dren: A review and algorithm. Pediatr Neurol 2010;43:377e90.

[12] Shonsted-Madsen U, Sehested P, Brask T. Benign intracranialhypertension caused by mastoiditis and lateral sinus obstruction:The value of computerized tomography in diagnosis. J LaryngolOtol 1984;98:395e8.

[13] Thapa R, Mallick D, Biswas B. Pseudotumor cerebri in childhoodhepatitis E virus infection. Headache 2009;49:610e1.

[14] Thapa R, Ghosh A, Mukherjee S. Childhood hepatitis A virusinfection complicated by pseudotumor cerebri. South Med J 2009;102:204e5.

[15] Tasdemir HA, Dilber C, Totan M, Onder A. Pseudotumor cerebricomplicating measles. Brain Dev 2006;28:395e7.

[16] Konrad D, Kuster H, Hunziker UA. Pseudotumor cerebri aftervaricella. Eur J Pediatr 1998;157:904e6.

[17] Lahat E, Leshem M, Barzilai A. Pseudotumor cerebri complicatingvaricella in a child. Acta Paediatr 1998;87:1310e1.