The Egyptian Journal of Radiology and Nuclear Medicine (2015) 46, 1129–1141

Egyptian Society of Radiology and Nuclear Medicine

The Egyptian Journal of Radiology andNuclearMedicine

www.elsevier.com/locate/ejrnmwww.sciencedirect.com

ORIGINAL ARTICLE

Assessment of the diagnostic reliability of brain CT

and MRI in pediatric epilepsy patients

Abbreviations: FCD, focal cortical dysplasia; AVM, arterio-venous

malformation; PVL, periventricular leukomalacia; TS, tuberous

sclerosis; CMV, cytomegalo virus; TB, tuberculosis; WM, white

matter; ADEM, acute disseminated encephalomyelitis; ADL,

adrenoleukodystrophy.* Corresponding author at: Department of Radiology, Sohag Faculty

of Medicine, Sohag University, Egypt. Tel.: +20 1224648517, +20

932307121.

E-mail address: m.elkousy@yahoo.com (M.H. Alam-Eldeen).

Peer review under responsibility of Egyptian Society of Radiology and

Nuclear Medicine.

http://dx.doi.org/10.1016/j.ejrnm.2015.07.0010378-603X � 2015 The Authors. The Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Mohamad Hasan Alam-Eldeen *, Nahla Mohamed Ali Hasan

Department of Diagnostic Radiology, Sohag Faculty of Medicine, Sohag University, Egypt

Received 27 April 2015; accepted 4 July 2015

Available online 26 July 2015

KEYWORDS

Pediatric epilepsy;

CT;

MRI

Abstract Objective: Analysis of the role of CT and MRI in evaluation of pediatric epilepsy and

their diagnostic utility to detect various pathologic entities at one center.

Patients and methods: From April 2012 to April 2014, 181 children with epilepsy were underwent

standardized CT and or MRI in the Department of Diagnostic Radiology at Sohag University

Hospitals. We retrospectively analyzed whether lesions were detected and whether were precisely

characterized by CT and or MRI.

Results: Of the 181 patients included in the study, abnormalities were detected in 132 patients

(73%) while normal brain images were found in 49 patients (27%). Of the 74 patients had CT only,

abnormalities were detected in 73% while 27% were normal. Of the 89 patients underwent MRI

only, abnormalities were detected in 74% while 26% were normal. Of the 18 patients underwent

CT and MRI, abnormalities were detected in 67% while 33% were normal.

Conclusion: Seizure disorders are among the most frequent neurologic problems that occur in

childhood. MRI is the technique of choice and is superior to CT in identifying underlying cause.

However, CT can serve as a screening modality and MRI would be performed to define abnormal-

ities found on CT or in negative CT studies.� 2015 The Authors. The Egyptian Society of Radiology and Nuclear Medicine. Production and hosting

by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://

creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Epilepsy is a chronic condition characterized by recurrentunprovoked seizures displaying the clinical manifestation of

excessive, abnormal, unprovoked, synchronous discharges ofneurons (1). Childhood epilepsies are a heterogeneous groupof conditions that differ in their diagnostic criteria and man-agement and have dramatically different outcomes. There are

different classification systems for epilepsy, commonly accord-ing either anatomy or etiology. According to etiology, epilepsyis classified into developmental, neoplastic, vascular, post trau-

matic, inflammatory or infective conditions (1,2).

Table 1 Positive and negative findings in the different imaging

modalities used.

Imaging modality No of patients +ve Findings �ve FindingsI. CT 74 54 (73%) 20 (27%)

II. MRI 89 66 (74%) 23 (26%)

III. CT & MRI 18 12 (67%) 6 (33%)

Total 181 132 (73%) 49 (27%)

1130 M.H. Alam-Eldeen, N.M.A. Hasan

The role of neuroimaging is to detect an underlying cerebrallesion(s) that may be related to the child’s seizure disorder that

is associated in some cases with neurodevelopmental impair-ment. Imaging is obtained to establish etiology, to provideprognosis, and to plan appropriate clinical care (3). CT can

detect some abnormalities and easily detects conditions

Fig. 1 FCD. Axial FLAIR (A), T1 pre contrast (B) and T1 post con

frontal lobe that is of abnormal high signal on FLAIR with blurred j

Fig. 2 Schizencephaly. (A) Patient with Rt sided open lip schizenceph

nipple sign is evident on (C). (D–F) Patient with Lt sided open lip schiz

seen on CT and MRI.

associated with calcification such as found in congenital infec-tions. MRI is fundamental in epilepsy imaging, providingstructural, anatomical and pathologic informations (4).

Recent advances in MRI, have increased our understandingof the underlying disease process which is reflected on manage-ment of epilepsy (5).

In this study we analyzed whether lesions were detected andwhether they were precisely characterized by CT and standard-ized MRI and in which cases CT is a sufficient imaging tool for

pediatric epilepsy at one center.

2. Patients and methods

This retrospective hospital-based study included 181 pediatricpatients (96 boys and 85 girls, mean age was 4 years and3 months, ranged from 1 month to 17 years) that were clini-

cally diagnosed as having epilepsy and referred to the

trast (C) showing a non enhancing focal cortical thickening at Rt

unction with WM.

aly. (B and C) Patient with Rt sided closed lip schizencephaly, the

encephaly, the ectopic gray matter lining the cleft margins is clearly

Fig. 3 Semilobar holoprosencephaly. (A and B) Axial and coronal T2 showing semilobar holoprosencephaly, the sylvian fissures lie

more anterior than usual.

Fig. 4 Subependymal heterotopia. CT scan shows multiple

heterotopic gray matter nodules along subependymal region of Lt

lateral ventricle.

Fig. 5 Pachygyria. CT scan shows b

Assessment of the diagnostic reliability of brain CT and MRI 1131

Department of Diagnostic Radiology at Sohag UniversityHospitals for imaging assessment in the period from April

2012 to April 2014. Of the 181 patients included in the study,74 patients referred for CT scan only, 89 patients for MRI onlyand 18 patients for CT scan and MRI. Children with intracra-

nial tumors and CNS postoperative cases were excluded fromthe study due to absence of operative and histopathologicaldata.

2.1. CT technique

Unenhanced and when necessary enhanced (intravenousadministration of 2 ml/kg of Ultravist 300 mg/ml) brain

CT-scans were obtained on GE Lightspeed ultra 8 slicescanner with a slice thickness 5–8 mm.

2.2. MRI technique

MRI was performed on a 1.5T unit (Philips-Acheiva,Netherlands) using dedicated head coil. The protocol included

the following sequences: axial T1-weighted, T2-weighted, and

ilateral frontal lobes pachygyria.

Fig. 6 Polymicrogyria. CT scan shows focal polymicrogyria at

Rt parietal lobe.

Fig. 7 Bilateral underdeveloped operculum. CT scan shows

marked bilateral underdeveloped operculum with exposure of

insula.

1132 M.H. Alam-Eldeen, N.M.A. Hasan

fluid-attenuated inversion recovery (FLAIR) images; coronal

T2-weighted and sagittal T1-weighted images with the follow-ing parameters; T1-weighted spin-echo (TR – 450 ms; TE –15 ms), T2-weighted spin-echo (TR – 5509 ms; TE – 110 ms),

FLAIR images (TR – 900 ms; TE 140 ms). All MR imagesare obtained at a 5-mm section thickness with a 1 mm intersec-tion gap.

In patients with the epileptogenic zone in the temporal lobe,coronal images were acquired perpendicular to the hippocam-pal axis. Contrast enhanced spin-echo T1-weighted images

were performed after the intravenous administration of mag-nifest 0.2 ml/kg body weight in indicated cases.

2.3. Image analysis

The CT and MR images were reviewed by two radiologists forinterpretation. They were classified as normal and abnormal,with attempt to reach to a histopathologic diagnosis.

Fig. 8 Hypoxic ischemia. Non enhanced CT shows diffuse bilateral h

demarcation and sparing of thalami.

3. Results

Of the 181 patients included in the study, abnormalities weredetected in 132 patients (73%) while normal brain images were

found in 49 patients (27%).Of the 74 patients underwent CT only, abnormalities

detected in 54 patients (73%) while 20 patients (27%) were nor-

mal and they did not undergo MRI. Of the 89 patients under-went MRI only, abnormalities detected in 66 patients (74%)while 23 patients (26%) were normal. Of the 18 patients

underwent CT and MRI, abnormalities detected in 12 patient(67%) while 6 patients (33%) were normal (Table 1).

In descending order of frequency, the abnormalities thatwas detected on either CT or MRI in the 132 patients was as

follows: cortical malformation disorders in 50 patients (37.9%), vascular lesions in 22 patients (16.6%), phakomatoses in20 patients (15%), infective and inflammatory lesions in 15

patients (11.4%), white matter disorders in 13 patients

ypodensity of cerebral hemispheres with loss of gray–white matter

Fig. 9 AVM. Axial T1 pre contrast (A), axial T1 post contrast (B), axial T2 (C) and coronal and sagittal T1 post contrast (D) showing

Lt parietal parasagittal AVM.

Assessment of the diagnostic reliability of brain CT and MRI 1133

(9.8%), encephalomalacia in 8 patients (6%) and mesial tem-poral sclerosis in 4 patients (3.3%).

Of the 50 patients with cortical malformation disorders;FCD was found in 11 patients, MRI only was done for 9patients while MRI and CT were done for 2 patients (none

was diagnosed on CT and all of them were diagnosed onMRI) (Fig. 1); schizencephaly was found in 9 patients, 4 wereunderwent CT only (3 open lip and 1 closed lip), 4 were under-went MRI only (2 open lip and 2 closed lip) and 1 patient had

CT and MRI and was diagnosed on both as open lip (Fig. 2);holoprosencephaly was found in 7 patients, CT only was donefor 3 (1 was alobar and 2 were semilobar), MRI only was done

for 3 (2 were semilobar and 1 was lobar) and CT and MRIwere done for 1 patient (diagnosed as semilobar on CTand MRI) (Fig. 3); heterotopia was found in 6 patients, 4

had MRI only and were diagnosed as nodular heterotopia, 2had CT and MRI (1 was diagnosed as subependymal hetero-topia on CT and MRI while one was diagnosed as nodular

heterotopia on MRI only and was not evident on CT)(Fig. 4); hemimegalencephaly was found on 5 patients, all of

them were underwent MRI only; pachygyria was found in 5patients and was focal on all, 2 had CT only and 3 had MRI

only (Fig. 5); polymicrogyria was found in 4 patients andwas focal in all of them, 3 had MRI only and 1 had bothCT and MRI and it was evident on both CT and MRI

(Fig. 6); bilateral underdeveloped operculum was found in 3patients all were underwent CT only (Fig. 7).

Vascular lesions were diagnosed in 22 patients; hypoxicischemic changes were found in 9 patients, 5 had CT only

and 4 had MRI only (Fig. 8); AVM was found in 4 patientsand was parenchymal in all cases, 2 were underwent CT onlyand 2 were underwent MRI only (Fig. 9); multifocal ischemia

due to vasculitis was found in 3 patients who were underwentMRI only (Fig. 10) and the diagnosis was supported by theclinical history; PVL was found in 3 patients, 2 had CT only

and 1 had MRI only (Fig. 11); hydranencephaly was found in3 patients who were underwent CT only (Fig. 12).

Phakomatoses were diagnosed in 20 patients; TS was found

in 14 patients, all of them were underwent CT only and theyhad cortical and subependymal calcified tubers with two of

Fig. 10 Multifocal ischemia in patient with nephrotic syndrome. Axial FLAIR (A), T1 (B), T2 (C) and DWI (D) showing bilateral

multifocal cortical ischemia.

Fig. 11 Periventricular leukomalacia. Non enhanced CT shows bilateral foci of PVL with reduced thickness of the white matter and mild

dilated ventricles posteriorly.

1134 M.H. Alam-Eldeen, N.M.A. Hasan

them had giant cell astrocytoma (Fig. 13); Sturge–Weber syn-drome was found in 6 patients, 2 had CT only, 3 had MRIonly, 1 had CT and MRI and was evident on both (Fig. 14).

Infective or inflammatory lesions were diagnosed in 15patients; congenital CMV was found in 7 patients, all of themwere underwent CT only and show diffuse cerebral atrophy,

ventriculomegaly and periventricular calcification (Fig. 15);

encephalitis was found in 6 patients, all of them were under-went MRI only and the diagnosis was supported by the clinicalhistory (Fig. 16), TB was found in 2 patients, one had CT only

and one had MRI only (Fig. 17) and the diagnosis was sup-ported by clinical history and laboratory findings.

White matter disorders were diagnosed in 13 patients;

diffuse WM dysmyelinating disorder was found in 6 patients

Fig. 12 Hydranencephaly. Non enhanced CT shows complete bilateral absence of cerebral parenchymas which are replaced by CSF.

Fig. 13 Tuberous sclerosis. (A and B) Non enhanced CT shows multiple calcified subependymal and cortical tubers. (C) Non enhanced

CT in a different patient with TS and Lt sided giant cell astrocytoma.

Fig. 14 Sturge Weber syndrome. Non enhanced CT (A), axial T2 (B) and axial FLAIR (V) clearly showing the characteristic cortical

calcification on CT and the gliosis and atrophy on MRI.

Assessment of the diagnostic reliability of brain CT and MRI 1135

and all had MRI only, ADEM was found in 3 patients and allhad MRI only (Fig. 18), ADL was found on 2 patients, 1 hadCT only and 1 had MRI only (Fig. 19), Alexander’s disease

was found in 2 patients and both had MRI only (Fig. 20).

Encephalomalacia was found in 8 patients, 5 were under-went CT only and 3 were underwent MRI only (5 were multi-focal with history of hypoxic ischemia and 3 were due to old

infarctions) (Fig. 21). Mesial temporal sclerosis was found in

Fig. 15 Congenital CMV infection. Non enhanced CT shows cerebral atrophy, ventriculomegaly and bilateral periventricular

calcification.

Fig. 16 Encephalitis. Axial FLAIR (A), T2 (B), DWI (C) and T1

post contrast (D) showing encephalitis along Rt fronto-parietal

lobes. Another smaller area also noted on Lt side.

1136 M.H. Alam-Eldeen, N.M.A. Hasan

4 patients, all of them had CT and MRI and were diagnosed

on MRI while they were not evident on CT (Fig. 22) (Tables2–4).

4. Discussion

Presence of seizures constitutes a symptom of an underlyingcentral nervous system disorder that requires thorough investi-

gations and management plan. The etiological factorsdiffer markedly in children (6). Congenital infection andperinatal ischemia are diagnostic considerations in neonates.

In infants below 1.5 years, mesial temporal sclerosis is not a

consideration. After 1.5 years, the yield of structural pathologyis highest (1).

Neuroimaging plays an essential role in noninvasively local-

izing epileptogenic foci in the pediatric population (1). CT candetect many structural abnormalities and offers some advan-tages with regard to identifying calcification as found incongenital infections, calcified tubers in TS and gyral calcifica-

tion in Sturge–Weber syndrome. In our study, 74 patientsunderwent CT only, lesions were identified in 54 patients(73%), 12 patients had cortical malformations disorders, 12

had vascular lesions, 16 had phakomatoses, 8 had infectivelesions, 1 had WM disease and 5 had encephalomalacia.

Being much more sensitive than CT, MRI is the technique

of choice in identifying the underlying cause in epilepsy (7,8).This is related to its high resolution structural imaging, supe-rior soft tissue contrast and multiplanar imaging capability

which allow visualization of epileptogenic substrates withgreater sensitivity and accuracy (9). In our study, 89 patientswere underwent MRI only, lesions were identified in 66patients (74%), cortical malformations disorders were found

in 31 patients, vascular lesions in 10, phakomatoses in 3, infec-tive lesions in 7, WM diseases in 12 and encephalomalacia in 3.

MRI has superiority on CT in the diagnosis of some causes

of epilepsy specially in patients with mesial temporal sclerosisand small FCD that will not be identified by CT (3,8). In ourstudy, 18 patients were underwent both CT and MRI, lesions

were identified in 12 patients (67%), 2 cases of FCD were diag-nosed by MRI and not identified on CT, one case diagnosed asopen lip schizencephaly by CT and MRI, one case diagnosedas semilobar holoprosencephaly by CT and MRI, 2 cases of

gray matter heterotopia; one was diagnosed as subependymalheterotopia on CT and MRI while the other one was diag-nosed as subcortical nodular heterotopia on MRI and was

not evident on CT, one case was diagnosed as focal polymicr-ogyria by CT and MRI, one case diagnosed as Sturge–Webersyndrome by CT and MRI with the calcification was evident

on CT, and 4 cases were diagnosed as mesial temporal sclerosisby MRI that were not detected by CT.

However, some, but not all, causes of pediatric epilepsy are

not detectable with conventional MRI (4,10). In our study, nolesions detected in 23 patients (26%) from the 89 patients

Fig. 17 Tuberculosis. Post contrast CT showing extra and intra-axial enhancing nodular lesions.

Table 2 Etiological entities diagnosed by CT and MRI.

Etiology No (%) On CT On MRI CT & MRI

I. Cortical malformation disorders 50 (37.9%)

A. Focal cortical dysplasia 11 – 9 2

B. Schizencephaly 9 4 4 1

C. Holoprosencephaly 7 3 3 1

D. Heterotopia 6 – 4 2

E. Hemimegalencephaly 5 – 5 –

F. Pachygyria 5 2 3 –

G. Polymicrogyria 4 – 3 1

H. Bilateral underdeveloped operculum 3 3 – –

II. Vascular 22 (16.6%)

A. Hypoxic ischemia 9 5 4 –

B. AVM 4 2 2 –

C. Vasculitis 3 – 3 –

D. Periventricular leukomalacia 3 2 1 –

E. Hydranencephaly 3 3 – –

III. Phakomatoses 20 (15%)

A. Tuberous sclerosis 14 14 – –

B. Sturge Weber syndrome 6 2 3 1

IV. Infective/inflammatory 15 (11.4%)

A. Congenital CMV 7 7 – –

B. Encephalitis 6 – 6 –

C. TB 2 1 1 –

V. White Matter disorders 13 (9.8%)

A. Diffuse WM disease 6 – 6 –

B. ADEM 3 – 3 –

C. Adrenoleukodystrophy 2 1 1 –

D. Alexander’s disease 2 – 2 –

VI. Encephalomalacia 8 (6%) 5 3 –

VII. Mesial temporal sclerosis 4 (3.3%) – – 4

Total 132 (100) 54 66 12

Assessment of the diagnostic reliability of brain CT and MRI 1137

underwent MRI only and no lesions detected in 6 patients(33%) from the 18 patients underwent both CT and MRI.

Cortical malformation disorders group is one of the com-monest causes for epilepsy in infants and children, accountingfor 10–50% of the pediatric epilepsy cases (11–13). It was

found in 50 patients (37.9%) in our study. It is important torecognize that in many lesions, CT changes are similar to thosefound in MRI studies (21). In our study, one case was

diagnosed as open lip schizencephaly on CT and MRI, onecase was diagnosed as semilobar holoprosencephaly on CT

and MRI, one case was diagnosed as subependymal hetero-topia on CT and MRI and one case was diagnosed as focalpolymicrogyria on CT and MRI.

CT scan however can fail to make the diagnosis in morethan 30% of affected patients (14). MRI allowed increasedrecognition and more accurate diagnosis (14,15). In our study,

Fig. 18 ADEM. Axial T2 (A) and FLAIR (B) showing bilateral multiple discrete patches of bright signals affecting gray and white

matter.

Fig. 19 Adrenoleukodystrophy. Non enhanced CT showing bilateral almost symmetrical white matter hypodensity at posterior parietal

lobes.

Fig. 20 Alexander’s disease. Axial T2 (A) and FLAIR (B) showing bilateral almost symmetrical white matter hyperintensity at a fronto-

parietal distribution.

1138 M.H. Alam-Eldeen, N.M.A. Hasan

Fig. 21 Encephalomalacia. (A) Non enhanced CT shows bilat-

eral multifocal areas of encephalomalacia with cerebral atrophy.

(B) Non enhanced CT of different patient with Lt fronto-parietal

lobes encephalomalacia.

Fig. 22 Mesial temporal sclerosis. Coronal T2 shows abnormal

bright signal at Rt hippocampus with dilated Rt temporal horn.

Table 3 Comparison of the diagnostic effectiveness between

CT and MRI in the 12 patients who were underwent both CT

and MRI and had shown +ve findings.

Etiology No +ve on CT +ve on MRI

Focal cortical dysplasia 2 0 2

Schizencephaly 1 1 1

Holoprosencephaly 1 1 1

Heterotopia 2 1 2

Polymicrogyria 1 1 1

Sturge Weber syndrome 1 1 1

Mesial temporal sclerosis 4 – 4

Total 12 5 12

Assessment of the diagnostic reliability of brain CT and MRI 1139

2 cases of FCD were diagnosed by MRI while they were notevident on CT, one case of subcortical nodular heterotopia

was diagnosed by MRI while it was not evident on CT.FCD is one of the most common forms of focal develop-

mental disorders constituting approximately 25% of patients

(13,16). On MRI, it is seen as focal cortical thickening that

Table 4 Categories of specific disease processes diagnosed by CT a

Etiology No Category 1

Schizencephaly 9 Open (6)

Holoprosencephaly 7 Alobar (1)

Heterotopia 6 Nodular (5)

ATS 14 Giant cell astrocyto

Encephalomalacia 8 Post hypoxia (5)

is hyperintense in T2 and FLAIR and hypointense in T1.The abnormal signal may extend to the underlying WM with

blurring of the gray–white matter interface. It is usually foundin frontal or temporal lobes, although it can occur in any loca-tion (10,17). It is infrequently associated with edema or gliosis

(18). In our study, from the 50 patients with cortical malforma-tion disorders, FCD was found in 11 patients (22%), all ofthem were in the frontal lobe and were associated with mild

blurring of underlying gray–white matter interface. No edemaor gliosis was found.

In schizencephaly, adjacent areas of polymicrogyria orpachygyria may be seen (14). It can be unilateral or bilateral

and tends to involve the insular, precentral and postcentralregions (13). Closed lip schizencephaly is characterized bythe presence of an indentation in the ependymal surface, the

so-called nipple sign (14). In our study, 3 cases were diagnosedas closed lip schizencephaly and all showed the nipple sign. Ofthe 6 cases diagnosed as open lip schizencephaly, one case was

detected in the precentral region and spared the insula.In holoprosencephaly, it is found that the sylvian fissures

were found more anterior than usual with respect to the frontof the brain (19). This is in agreement with our results as the

sylvian fissures were present in an abnormal anterior location.The most common type of heterotopic gray matter is the

subependymal nodular type. It can be isolated or associated

with other anomalies of the brain (20). It can be associatedwith mesial temporal sclerosis (21). Mild ventricular dilationmight be seen (18,22). Single case was diagnosed as subependy-

mal heterotopia in our study and it was associated with mildventricular dilatation but no other anomalies were detected.

In hemimegalencephaly, there is unilateral focal or diffuse

dysplastic cortical thickening. Commonly, there is indistinctgray–white matter junction, ipsilateral WM myelination disor-ders, heterotopias, lissencephaly and polymicrogyria (10,23).In the 5 cases of hemimegalencephaly in our study, dysplastic

cortex of the contralateral cerebral hemisphere was detected inone case.

nd or MRI.

Category 2 Category 3

Closed (3)

Semilobar (5) Lobar (1)

Subependymal (1)

ma (2)

Post infarction (3)

1140 M.H. Alam-Eldeen, N.M.A. Hasan

Pachygyria or incomplete lissencephaly represents a fewlarge, thick and smooth gyri, with shallow sulci. It can be uni-lateral or bilateral, focal or diffuse (24,25). In the 5 cases of

pachygyria in our study, it was focal in all cases and was accu-rately diagnosed by CT in 2 cases and by MRI in 3 cases.

Polymicrogyria may be unilateral or bilateral, frontal, fron-

toparietal, perisylvian, parieto-occipital or generalized (26,27).In localized form, the remaining cortical gray matter may beabnormal (28). The 4 patients with polymicrogyria in our

study were of unilateral frontoparietal pattern and were notassociated with other cortical abnormality. Poly-microgyriamay occur adjacent to the borders of schizencephaly (14). Inthe 9 cases with schizencephaly in our study, there was no asso-

ciated polymicrogyria.In bilateral underdeveloped operculum, CT shows symmet-

rical bilateral enlarged sylvian fissures, underdevelopment of

the operculum, exposed insula, with or without normallyformed insular gyri (29–31). The 3 cases of bilateral underde-veloped operculum in our study showed bilateral enlarged syl-

vian fissures, underdeveloped operculum and exposed insula.Normal gyral pattern was seen in 2 cases while polymicrogyriapattern was seen in one case.

In periventricular leukomalacia, there are hyperintense fociin the periventricular WM on T2 and FLAIR associated withreduced thickness predominantly in the periatrial regions withcompensatory focal ventricular enlargement (32). Our 3 cases;

2 of them diagnosed on CT; showed the typical describedpattern.

The best diagnostic clue in tuberous sclerosis is calcified

subependymal nodules and subependymal giant cell astrocy-toma. Cortical, subcortical tubers are present in 70–95%, witha predilection of the frontal and parietal lobes (1,33). Calcified

subependymal and cortical tubers were found in the 14 cases oftuberous sclerosis in our study and subependymal giant cellastrocytoma was found in 2 cases.

On CT, Sturge–Weber syndrome has characteristic gyri-form calcifications, more commonly affect the parieto-occipital areas, with underlying focal brain atrophy.Leptomeningeal enhancement is well seen on MRI and the

areas of gliosis are well seen on T2 and FLAIR (10,34). Inthe 6 cases diagnosed in our study, the calcification was evidentby CT, gliosis was evident by MRI and focal atrophy was evi-

dent by CT and MRI.CT is highly sensitive for the depiction and localization of

calcification in cases of congenital CMV infection.

Calcification is commonly thick and chunky, commonly seenin periventricular regions but may occur in basal ganglia andbrain parenchyma. Congenital CMV infection is also charac-terized by cerebral atrophy and ventriculomegaly (35). A vari-

ety of associated migrational abnormalities like lissencephaly,pachygyria, and polymicrogyria have been reported (36). The7 cases diagnosed as congenital CMV infection in our study

were underwent CT and showed typical thick periventricularcalcification, cerebral atrophy and ventriculomegaly with noany associated anomalies.

Adrenoleukodystrophy is characterized by symmetric WMdemyelination in the peritrigonal regions that extends acrossthe corpus callosum splenium. It may then spread outward

and cephalad as a confluent lesion. The subcortical WM is rel-atively spared in the early stage (37). The 2 cases in our studyhad a symmetrical peritrigonal distribution.

Alexander disease has a predilection for the frontal lobesWM early in its course then progresses posteriorly to the pari-etal WM and internal and external capsules. The subcortical

WM is affected early in the disease course (37). The 2 casesin our study had an almost symmetrical bilateral fronto-parietal distribution.

The major findings of mesial temporal sclerosis on MRI arehippocampal atrophy, abnormal T2 hyperintense signal inmesial temporal region and dilatation of the ipsilateral tempo-

ral horn. Secondary findings include loss of internal architec-ture, atrophy of the ipsilateral fornix and mammillary body,and temporal lobe volume loss (1,10). The 4 cases of mesialtemporal sclerosis in our study were diagnosed on MRI and

not on CT, they showed abnormal hyperintense signal onmesial temporal region with mild dilatation of adjacent tempo-ral horn.

5. Conclusion

Seizure disorders are among the most frequent neurologic

problems that occur in childhood. Neuroimaging plays anessential role in noninvasively localizing epileptogenic foci.MRI is the technique of choice and is superior to CT in iden-

tifying underlying cause. However, CT is more widely avail-able than MRI, less expensive, and can diagnose manyabnormalities. So, CT can serve as a screening modality and

MRI would be performed to define abnormalities found onCT or in �ve CT studies.

5.1. Limitations

Our study has limitations. First, CT and MRI were not done inall patients, so the precise sensitivity of CT and MRI in diag-nosis could not be gained. Second, in the patients who under-

went CT only and diagnosed as normal, MRI was notperformed to confirm absence of lesions.

Conflict of Interest

We have no conflict of interest to declare.

References

(1) Lai V, Mak HKF. Neuroimaging in paediatric epilepsy. Med Bull

2009;14(5).

(2) Jackson GD et al. Temporal lobe epilepsy. In: Kuzniecky RI,

Jackson GD, editors. Magnetic resonance in epilepsy, 2nd ed.;

2005. p. 99–220.

(3) Gaillard WD, Chiron C, Cross H, Harvey S, et al. Guidelines for

imaging infants and children with recent-onset epilepsy. Epilepsia

2009;50(9):2147–53.

(4) Duncan JS. The epilepsies. Brain mapping: the disorders; 2000. p.

317–55.

(5) Saneto RP, Wyllie E. Surgically treatable epilepsy syndromes in

infancy and childhood. In: Miller JW, editor. Epilepsy surgery –

principles and controversies; 2006. p. 121–41.

(6) Veena K, Ghai OP, Ghai OP, et al. Epilepsy. Essential pediatrics.

CBS; 2006. p. 509–14.

(7) Urbach H, Hattingen J, von Oertzen J, Luyken C, et al. MR

imaging in the presurgical workup of patients with drug-resistant

epilepsy. AJNR 2004;25:919–26.

Assessment of the diagnostic reliability of brain CT and MRI 1141

(8) Guissard G, Damry N, Dan B, et al. Imaging in pediatric

epilepsy. Arch pediatr 2005;12(3):337–46.

(9) Lai V, Mak HKF, Yung ADW, Ho WH, Hung KN.

Neuroimaging techniques in epilepsy. Hong Kong Med J

2010;16(4).

(10) Rastogi S, Lee C, Salamon N. Neuroimaging in pediatric

epilepsy: a multimodality approach. Radiographics

2008;28(4):1079–96.

(11) Barkovich AJ et al. Classification system for malformations of

cortical development: update 2001. Neurology 2001;57:

2168–78.

(12) Barkovich AJ, Raybaud CA. Malformations of cortical develop-

ment. Neuroimaging Clin N Am 2004;14:401–23.

(13) Pang T, Atefy R, Sheen V. Malformations of cortical develop-

ment. Neurologist 2008;14(3):181–91.

(14) Andrade CS, Leite CC. Malformations of cortical development.

Current concepts and advanced neuroimaging review. Arq

Neuropsiquiatr 2011;69(1):130–8.

(15) Leite CC, Lucato LT, Sato JR, Valente KD, et al. Multivoxel

proton MR spectroscopy in malformations of cortical develop-

ment. AJNR 2007;28:1071–5.

(16) Barkovich AJ, Kuzniecky RI, Jacson GD, et al. A developmental

and genetic classification for malformation of cortical develop-

ment. Neurology 2005;65:1873–87.

(17) Sisodiya SM, Fauser S, Cross JH, Thom M. Focal cortical

dysplasia type II: biological features and clinical perspectives.

Lancet Neurol 2009;8:830–43.

(18) Abdel Razek AAK, Kandell AY, Elsorogy LG, Elmongy A,

Basett AA. Disorders of cortical formation: MR imaging features.

AJNR 2009;30:4–11.

(19) Barkovich AJ, Simon EM, Clegg NJ, Kinsman SL, et al. Analysis

of the cerebral cortex in holoprosencephaly with attention to the

sylvian fissures. AJNR 2002;23:143–50.

(20) Barkovich AJ. Morphologic characteristics of subcortical hetero-

topia: MR imaging study. AJNR 2000;21:290–5.

(21) Sisodiya SM. Malformations of cortical development: burdens

and insights from important causes of human epilepsy. Lancet

Neurol 2004;2:29–38.

(22) Barkovich J. Morphologic characteristics of subcortical hetero-

topia: MR imaging study. AJNR 2000;21:290–5.

(23) Yagishita A, Arai N, Tamagawa K, Oda M.

Hemimegalencephaly: signal changes suggesting abnormal myeli-

nation on MRI. Neuroradiology 1998;40:734–8.

(24) Kuzniecky RI, Jackson GD. Magnetic resonance in epilepsy. 2nd

ed. Oxford: Elsevier; 2005.

(25) Adachi Y, Poduri A, Kawaguch A, Yoon G, et al. Congenital

microcephaly with a simplified gyral pattern: associated findings

and their significance. AJNR 2011;32:1123–9.

(26) Leventer RJ, Jansen A, Pilz DT, Stoodley N, et al. Clinical and

imaging heterogeneity of polymicrogyria: a study of 328 patients.

Brain 2010;133:1415–27.

(27) Im K, Pienaar R, Paldino MJ, Gaab N, et al. Quantification and

discrimination of abnormal sulcal patterns in polymicrogyria.

Cereb Cortex 2012.

(28) Oliveira PP, Valente KD, Shergill SS, Leite CC, et al. Cortical

thickness reduction of normal appearing cortex in patients with

polymicrogyria. J Neuroimaging 2010;20:46–52.

(29) Kuzniecky R, Andermann FCBPS Study Group. The congenital

bilateral perisylvian syndrome: imaging findings in a multicenter

study. AJNR 1994;15:139–44.

(30) Chen CY, Zimmerman RA, Faro S, Parrish B, et al. MR of the

cerebral operculum: abnormal opercular formation in infants and

children. AJNR 1996;17:1303–11.

(31) Poduri A, Chitsazzadeh V, D’Arrigo S, et al. The syndrome of

perisylvian polymicrogyria with congenital arthrogryposis. Brain

Dev 2009;32:550–5.

(32) Kwee RM, Kwee TC. Virchow-robin spaces at MR imaging.

Radiographics 2007;27:1071–86.

(33) Baron Y, Barkovich AJ. MR imaging of tuberous sclerosis in

neonates and young infants. AJNR 1999;20:907–16.

(34) Guermazi A, De Kerviler E, Zagdanski AM, Frija J. Diagnostic

imaging of choroid plexus disease. Clin Radiol 2000;55:503–16.

(35) Fink KR, Thapa MM, Ishak GE, Pruthi S. Neuroimaging of

pediatric central nervous system cytomegalovirus Infection.

Radiographics 2010;30:1779–96.

(36) Van der Knaap MS, Vermeulen G, Barkhof F, Hart AAM, et al.

Pattern of white matter abnormalities at MR imaging: use of

polymerase chain reaction testing of guthrie cards to link pattern

with congenital cytomegalovirus infection. Radiology

2004;230(2):529–36.

(37) Cheon JE, Kim IO, Hwang YS, Kim KJ, et al. Leukodystrophy

in children: a pictorial review of MR imaging features.

Radiographics 2002;22:461–76.