ORIGINAL RESEARCH

PEDIATRICS

When Should a Brain MRI Be Performed in Children with

New-Onset Seizures? Results of a Large Prospective Trial
R. Hourani, W. Nasreddine, M. Dirani, G. Hmaimess, S. Sabbagh, O. El Tourjuman, J. Wazne, H. Toufaili,
N. AlArab, M. El Dassouki, and A. Beydoun

ABSTRACT
BACKGROUND AND PURPOSE: There is a paucity of data regarding the incidence of structural brain lesions in children with new-onset
unprovoked seizures. Our aim was to determine the frequencies and types of epileptogenic lesions detected on a dedicated epilepsy
protocol MR imaging according to age group, the presence of developmental delay, and the number and types of seizures.

MATERIALS AND METHODS: Consecutive children between 6 months and 18 years of age with new-onset unprovoked seizures
were included. The frequencies and types of epileptogenic lesions were determined and then stratified according to sex, age
groups, the presence of developmental delay, and the number and types of seizures at presentation. Multivariate analysis was used
to identify variables significantly associated with the presence of epileptogenic lesions.

RESULTS: One thousand children were included. An epileptogenic lesion was identified in 26%, with malformations of cortical develop-
ment being the most common lesion (32%), followed by hypoxic-ischemic injury (20%) and vascular etiologies (16%). Univariate analysis
showed a significant increase in the frequency of epileptogenic lesions with decreasing age, the presence of developmental delay, and
the number and types of seizures at presentation. The presence of developmental delay and seizure type at presentation remained sig-
nificant in a multivariate analysis.

CONCLUSIONS: We documented a relatively high rate of epileptogenic lesions in children with new-onset seizures, with the pres-
ence of developmental delay and specific seizure types being associated with a higher likelihood of detecting an epileptogenic
lesion on neuroimaging. This study fulfills the requirements of the study design recommended by the Practice Committee of the
American Academy of Neurology, and we hope that our results will assist the relevant societies and committees in formulating
neuroimaging guidelines for children with new-onset seizures.

ABBREVIATIONS: DD ¼ developmental delay; MCD ¼ malformations of cortical development; MTS ¼ mesial temporal sclerosis; NCS ¼ neurocutaneous
syndromes; PVL ¼ periventricular leukomalacia

A brain MR imaging is useful in the work-up of patients with

new-onset seizures because it can help define the electroclin-
ical syndrome, identify surgically remediable lesions, and assist in
establish the diagnosis of epilepsy in patients presenting with a
single unprovoked seizure.3
There is a paucity of data regarding the frequency of structural
predicting medical refractoriness.1,2 In addition, according to the brain lesions in children presenting with new-onset unprovoked
new proposed definition of epilepsy, a brain MR imaging may seizures. Etiologically related neuroimaging abnormalities were
identified in 13%–18% of such children, but those studies have sev-
Received December 21, 2020; accepted after revision April 5, 2021. eral methodologic drawbacks, including the acquisition of head
From the Departments of Radiology (R.H., N.A.) and Neurology (W.N., M.D., A.B.),
American University of Beirut Medical Center, Beirut, Lebanon; Department of
CTs, non-epilepsy protocol brain MRIs, and selection biases.4-6
Pediatrics (G.H.), St. George Hospital Medical University Center, University of The practice parameter issued in 2010 and reaffirmed in 2017
Balamand, Beirut, Lebanon; Department of Pediatrics (S.S.), Hotel Dieu de France
Hospital, Beirut, Lebanon; Department of Neurology (O.E.T., J.W., M.E.D.), Rafic
by the American Academy of Neurology7 determined that there
Hariri University Hospital, Beirut, Lebanon; and Department of Neurology (H.T.), was insufficient evidence to support a recommendation for rou-
Labib Medical Center, Beirut, Lebanon.
tine neuroimaging of children with a first afebrile seizure. To
R. Hourani and W. Nasreddine contributed equally to this work.
Data are available on reasonable scientific request.
overcome the shortcomings of prior studies and to generate de-
Please address correspondence to Ahmad Beydoun, MD, Professor of Neurology, finitive evidence regarding the value of neuroimaging studies in
American University of Beirut Medical Center, PO Box 11-0236, Riad El-Solh 1107 the pediatric population with new-onset seizures, a call was made
2020, Beirut, Lebanon; e-mail: [email protected]
for prospective data to be collected in sufficiently large samples,
http://dx.doi.org/10.3174/ajnr.A7193 allowing adequate statistical power to provide precise estimates
AJNR Am J Neuroradiol 42:1695–1701 Sep 2021 www.ajnr.org 1695
with narrow confidence intervals.7 In addition, the American Assessment of Intellectual or Global Developmental Delay
Academy of Neurology practice parameter stressed the impor- All children were evaluated for the presence and severity of DD.
tance of stratifying patients by age groups and including consecu- Children younger than 6 years of age were assessed with the
tive children for the results to be accurate and generalizable.7 Denver Development Screening Test.15 Older children were eval-
The primary aim of this study was to follow the recommenda- uated according to the Diagnostic and Statistical Manual of
tions of the American Academy of Neurology by prospectively Mental Disorders criteria that stratify intellectual disability into
assessing the frequencies and types of epileptogenic lesions in a mild, moderate, severe, and profound on the basis of deficits in
large cohort of consecutive children with new-onset seizures eval- intellectual functioning as well as difficulties in conceptual, social,
uated with a dedicated epilepsy protocol brain MR imaging. The and practical areas of living.16 For our analysis, we included only
secondary aims were to determine the yields and types of lesions 3 groups of DD (mild, moderate, or severe) by combining chil-
according to sex, the presence and severity of development delay dren with severe and profound delays into a single category.
(DD), as well as the number and types of seizure at presentation.
Classification of Seizure Types
MATERIALS AND METHODS The seizures were classified according to the recent International
Study Design and Patient Characteristics League Against Epilepsy operational classification.17,18 Children
Consecutive children between 6 months and 18 years of age were stratified into the following groups based on a detailed
diagnosed with $1 unprovoked seizure between November description of the seizure semiologies experienced at the time of
2010 and April 2017 were included in this study. Those children initial evaluation:
participated in an ongoing centralized prospective study evalu- Group 1: Children with epileptic spasms in clusters or fre-
ating patients with new-onset seizures. The details of this study quent tonic or atonic seizures
were previously reported.8 Group 2: Children with focal-onset seizures (focal aware or
focal impaired awareness) with or without focal-to-bilateral
Brain MRI and Classification of Neuroimaging Findings tonic-clonic seizures
Brain MRIs were obtained from a 1.5 or 3T scanner (Ingenia; Group 3: Children with unknown-onset tonic-clonic seizures
Phillips Healthcare) using an imaging-acquisition protocol that Group 4: Children with frequent absence seizures and/or
included 3D T1 (1 mm slice thickness) and 3D fast fluid-attenu- myoclonic seizures with or without generalized-onset tonic-
ated inversion recovery (FLAIR; 0.9 or 1 mm slice thickness) of clonic seizures
the whole brain with multiplanar reconstruction, axial and coro- Group 5: Children with unclassified seizure types. This cate-
nal inversion recovery (2 mm slice thickness), axial T2 TSE and gory included children who experienced what used to be labeled
T2 FFE (4 mm slide thickness) and axial diffusion weighted “dialeptic seizures” semiologically characterized by a loss of
images (4-5 mm slice thickness). The 3D images were obtained awareness and motionlessness that did not allow a definite dis-
with no interslice gap. This protocol satisfies all the recommenda- tinction based on the semiologic description alone between ab-
tions of the recently published Harmonized Neuroimaging of sence seizures and focal impaired-awareness seizures.
Epilepsy Structural Sequences (HARNESS-MRI)9 except for the
lack of acquisition of high in-plane resolution 2D coronal T2- Ethics Approval
weighted sequences using submillimetric voxel resolution. This study was approved by the American University of Beirut
The MRIs were interpreted by a neuroradiologist blinded to Medical Center institutional review board, and all parents signed
the clinical data with vast experience in the neuroimaging of an informed consent form. Additionally, children between 7 and
patients with epilepsy. 17 years of age signed an assent form.
MR imaging findings were classified as epileptogenic on the ba-
sis of previously published criteria.10-12 Epileptogenic lesions were Statistical Analyses
classified into the following categories: malformations of cortical We calculated the percentage of children with an epileptogenic
development (MCD), mesial temporal sclerosis (MTS), hypoxic- lesion and compared the frequencies and types of lesions accord-
ischemic injury (moderate or severe periventricular leukomalacia ing to age groups (0–2 years, 2–5 years, 5–10 years, 10–15 years,
[PVL] or hypoxic brain injury), vascular lesions, tumoral, neurocu- 15–18 years), DD, and types and number of seizures at baseline.
taneous syndromes (NCS), metabolic disorder, and others (eg, For continuous variables, descriptive statistics, including
postinfectious or posttraumatic encephalomalacia with gliosis, leu- mean, median, range, percentage, and 95% confidence interval
kodystrophy, and large arachnoid cysts exerting mass effect). were calculated. Statistical analyses were performed using the x 2
The cases of PVL were graded according to the following scoring test or Fisher exact test for categoric variables. Significant P values
system:13 severe PVL, diffuse white matter signal abnormality with were set at ,.05. Variables that showed a significant association
cystic changes; moderate PVL, diffuse white matter abnormality with- with the presence of an epileptogenic lesion in univariate analyses
out cystic changes; and mild PVL, isolated white matter abnormality. were entered into a multivariate model.
We considered only moderate or severe PVL as epileptogenic.14 In addition, a recursive partition analysis was performed to
MR imaging abnormalities consisting of isolated subcortical identify variables associated with higher or lower probabilities of
lesions or abnormal signal, nonspecific white matter hyperinten- detecting epileptogenic lesions. For this analysis, we used the x 2
sities, mild PVL, hydrocephalus, and brain atrophy were consid- Automatic Interaction Detector with cross-validation. At each
ered incidental findings. step, the x 2 Automatic Interaction Detector algorithm chooses
1696 Hourani Sep 2021 www.ajnr.org
Table 1: Demographic variables of the 1000 children included in underwent a non-epilepsy protocol brain MR imaging, and 3
the study children in whom a brain MR imaging was contraindicated.
Demographics Therefore, a total of 1000 children (boys = 58.1%, girls = 41.9%)
Age with a mean age of 7.8 years (range, 6 months to 17.9 years) were
Mean 7.8 yr included. The number of seizures at initial evaluation, seizure
Range 6 mo to 17.9 yr
Sex types, and psychomotor development are shown in Table 1.
Male/female 581/419
No. of seizures (%) MR Imaging Findings
1 seizure 315 (31.5%) Epileptogenic lesions were detected on the brain MR imaging of
Multiple seizures 685 (68.5%) 260 children (26%; 95% CI, 23.4%–28.8%). The frequencies of
Seizure type (No.) (%) epileptogenic lesions stratified by etiologic categories are shown
Focal-onset seizures 484 (48.4%)
Unknown-onset tonic-clonic seizures 260 (26.0%) in Fig 1. The most common type of epileptogenic lesion was
Absence and/or myoclonus with or 130 (13.0%) MCD detected in 84 children. The most frequent abnormalities
without generalized tonic-clonic seizures in this group consisted of focal cortical dysplasia (47.6%), fol-
Spasms, tonic or atonic seizures 93 (9.3%) lowed by polymicrogyria (22.6%), heterotopia (9.5%), multiple
Unclassified 33 (3.3%)
congenital malformations (8.3%), lissencephaly (7.1%), holopro-
Psychomotor development (No.) (%)
Normal 777 (77.7%) sencephaly (2.4%), and septo-optic dysplasia (2.4%). Evidence of
Delay 223 (22.3%) a hypoxic-ischemic injury was detected in 19.6% of children with
Mild DD 70 (7.0%) etiologically relevant lesions on neuroimaging. Epileptogenic
Moderate DD 63 (6.3%) lesions associated with a vascular etiology were detected in 16.2%
Severe/profound DD 90 (9.0%)
of children. Most (81%) had evidence of a prior ischemic infarc-
tion involving the cortex, 14% had a cavernoma, and 5% had an
arteriovenous malformation. The next most common type of epi-
leptogenic lesion was MTS, accounting for 7.7% of cases, followed
by tumoral etiologies and NCS, each diagnosed in 5.8% of chil-
dren. The tumors consisted of 8 neuroglial tumors, 4 infiltrative
astrocytomas, 2 hypothalamic hamartomas, and 1 epidermoid
cyst. Twelve of the 15 children with NCS were diagnosed with
tuberous sclerosis, with the remaining 3 diagnosed with neurofi-
bromatosis. The rest of the identified epileptogenic lesions con-
sisted of posttraumatic or postinfectious encephalomalacia with
cortical gliosis (n ¼ 14), metabolic disorder (n ¼ 9), leukodystro-
phy (n ¼ 6), and large arachnoid cysts exerting mass effect (n ¼ 4).

Frequencies of Epileptogenic Lesions According to Sex

The frequencies of epileptogenic lesions were not statistically sig-
nificant between boys (26%) and girls (26%).

Frequencies of Epileptogenic Lesions According to the

Number of Seizures at Presentation
At the initial evaluation, 315 children (31.5%) presented with a
single seizure. Those who presented with $2 seizures were signif-
icantly more likely to have an epileptogenic lesion (193/685,
28.2%; 95% CI, 24.9%–31.7%) compared with those with a single
FIG 1. Frequencies of epileptogenic lesions stratified by etiologic
seizure (67/315, 21.3%; 95% CI, 17.1%–26.1%; P ¼ .021). There
categories.
was, however, no significant difference in the types of identified
epileptogenic lesions between those 2 groups.
the independent variable that has the strongest interaction with
the dependent variable using P values with a Bonferroni correc-
Frequencies of Epileptogenic Lesions According to Age
tion as splitting criteria. The final result is a decision tree with
Groups
various nodes that can be used to predict the probability of There was a gradual and significant reduction in the frequencies
detecting an epileptogenic lesion in each subgroup. of epileptogenic lesions with ascending age groups (Table 2)
(P , .001). In addition, the predominant subtype of epileptogenic
RESULTS lesions varied according to age groups. For instance, hypoxic-
Patient Demographics ischemic lesion, which was the most common substrate in chil-
Of the 1160 consecutive children enrolled in this study, 160 were dren younger than 2 years of age and accounting for 31.7% of epi-
excluded for the following reasons: 94 because a brain MR imag- leptogenic lesions in that age group, gradually declined to
ing was not yet performed for financial or other reasons, 63 who account for 15.0% of lesions in the 5–10 year age group and 0%
AJNR Am J Neuroradiol 42:1695–1701 Sep 2021 www.ajnr.org 1697
Table 2: Frequencies of epileptogenic lesion subtypes stratified by age groupsa
0–2 Years (n = 169) 2–5 Years (n = 189) 5–10 Years (n = 281) 10–15 Years (n = 257) 15–18 Years (n = 104)
Epileptogenic lesions 82 (48.5%) 57 (30.2%) 60 (21.4%) 47 (18.3%) 14 (13.5%)
MCD 22 (26.8%) 20 (35.1%) 24 (40.0%) 18 (38.3%) 0 (0.0%)
Hypoxic-ischemic 26 (31.7%) 14 (24.6%) 9 (15.0%) 2 (4.3%) 0 (0.0%)
Vascular 14 (17.1%) 10 (17.5%) 6 (10.0%) 9 (19.1%) 3 (21.4%)
MTS 2 (2.4%) 1 (1.8%) 5 (8.3%) 9 (19.1%) 3 (21.4%)
NCS 8 (9.8%) 1 (1.8%) 4 (6.7%) 1 (2.1%) 1 (7.1%)
Tumoral 0 (0.0%) 3 (5.3%) 4 (6.7%) 4 (8.5%) 4 (28.6%)
Other 10 (12.2%) 8 (14.0%) 8 (13.3%) 4 (8.5%) 3 (21.4%)
a
95% CI for the percentages of epileptogenic lesions in the 0–2 year, 2–5 year, 5–10 year, 10–15 year, and 15–18 year age groups were 41.1%–56.0%, 24.1%–37%, 17.0%–
26.5%, 14.0%–23.5%, and 8.2%–21.3%, respectively.

Table 3: Frequencies of epileptogenic lesions stratified by analyses ($1 seizure on presentation, age groups, presence of
seizure type at presentation DD, and seizure types) as independent variables showed that
Epileptogenic only the presence of DD (odds ratio ¼ 4.3, P , .001) and seizure
Seizure Types Lesions
types on presentation (odds ratio ¼ 2.3, P , .001) remained sig-
Absence and/or myoclonus with or without 8 (6.2%)
nificant. The frequencies of epileptogenic lesions stratified by sei-
generalized tonic-clonic seizures (n ¼ 130)
Unknown-onset tonic-clonic seizures (n ¼ 260) 38 (14.6%) zure types and the presence or absence of DD are shown in Fig 2.
Unclassified (n ¼ 33) 5 (15.2%) The recursive analysis identified the same 2 variables (the
Focal-onset seizures (n ¼ 484) 148 (30.6%) presence or absence of DD and seizure type) that partitioned the
Spasms, tonic or atonic seizures (n ¼ 93) 61 (65.6%) patients into a decision tree with 4 groups (Fig 3). The highest
yield of detecting epileptogenic lesions (63.0%) was in the group
in the 15–18 year age group (Table 2). On the other hand, there of children with DD who presented with focal-onset seizures or
was a gradual increase in the frequencies of MTS and tumoral eti- with epileptic spasms or tonic or atonic seizures. The frequency
ologies with ascending age groups, peaking in the 15–18 year age of epileptogenic lesions in children without DD who presented
group (Table 2). Other types of epileptogenic lesions did not with the same seizure types was 24.8%. The corresponding yields
show an apparent age-related pattern. However, children with for children with and without DD who presented with other sei-
MCD presented with seizures before 15 years of age. zure types were 36% and 8.8%, respectively.

Frequencies of Epileptogenic Lesions According to the

Presence of DD
DISCUSSION
Two hundred twenty-three children had evidence of DD at the
This is the largest study that prospectively evaluated brain MR
initial evaluation, with a significant difference in frequency across
imaging findings in consecutive children with new-onset unpro-
the age groups (P , .001). DD was highest in the 0–2 year group
voked seizures using a dedicated imaging protocol. We docu-
(60.9%), followed by the 2–5 year group (29.1%), the 5–10 year
mented that the yield of identifying epileptogenic lesions in that
group (12.8%), the 10–15 year group (8.9%), and finally the 15–
patient population is 26% (95% CI, 23.4%–28.8%).
18 year group (5.8%).
Three previous studies identified etiologically related
The frequency of epileptogenic lesions was significantly
neuroimaging abnormalities in 13%–18% of children with new-
higher in children with DD (127/223, 57.0%; 95% CI, 50.4%–
onset seizures.4-6 Those studies had several methodologic short-
63.3%) compared with those with normal development (133/777,
comings, including evaluating children with a mixture of head CT
17.1%; 95% CI, 14.6%–19.9%; P , .001).There was also a signifi-
and non-epilepsy protocol brain MRIs,4,5 selection bias because
cant increase in the frequencies of epileptogenic lesions as the se-
neuroimaging studies were not systematically performed but only
verity of DD worsened (P , .001).
as clinically indicated,4,5 and exclusion of children with DD.6 For
Frequencies of Epileptogenic Lesions According to the instance, epileptogenic lesions were detected in 18% of 411 children
Type of Seizures Groups who presented with a first afebrile seizure.4 In that study, neuroi-
There was a significant difference in the frequencies of identified maging was only performed in 53% of children, with most studies
epileptogenic lesions according to the seizure types at presentation consisting of head CTs. In a community-based study of 613 chil-
(P , .001, Table 3). The lowest yield of detecting a lesion was in the dren with newly diagnosed epilepsy, neuroimaging studies (approx-
group of patients with absences and/or myoclonus with or without imately two-thirds had a standard brain MR imaging, and one-
generalized tonic-clonic seizures (6.2%), and the highest yield was in third underwent a head CT) were obtained in 80%, with etiologi-
children who presented with epileptic spasms in clusters or frequent cally relevant abnormalities detected in 13%.5 A subsequent pro-
tonic or atonic seizures (65.6%). The frequency of epileptogenic spective study performed brain MR imaging in 281/349 children
lesions in children with focal-onset seizures was 30.6% (Table 3). with a first-recognized seizure.6 Significant abnormalities or those
potentially related to seizures were identified in 14% of those chil-
Multivariate Analysis dren. The relatively low yield of epileptogenic lesions in that study
A logistic regression analysis with an epileptogenic lesion as the can be explained by the fact that children younger than 6 years of
dependent variable and all significant variables in the univariate age and those with moderate or severe DD were excluded.6
1698 Hourani Sep 2021 www.ajnr.org
FIG 2. Percentages of children with epileptogenic lesions stratified according to seizure types and the presence or absence of developmental
delay. GTC indicates generalized tonic-clonic seizures.

FIG 3. Recursive partition analysis stratified children into 4 groups based only on the presence of DD and seizure types. GTC indicates general-
ized tonic-clonic seizures.

The most common epileptogenic lesion in our study was concordant with those of a smaller study conducted in children
MCD, followed by hypoxic-ischemic injuries and vascular younger than 2 years of age with newly diagnosed epilepsy in
lesions. Because of the large number of children enrolled, our whom the most common pathologic substrate was developmental
study is the first to stratify the pathologic substrate on brain MR brain malformations.19 We also found that MCD was one of the
imaging according to age groups. We found that in children 2 most common epileptogenic lesions in the 2–15 year age group,
years of age and younger, the most common underlying etiology while MTS was most frequent in the 10–18 year age group pre-
was hypoxic-ischemic injury and MCD. Those results are senting with new-onset seizures.
AJNR Am J Neuroradiol 42:1695–1701 Sep 2021 www.ajnr.org 1699
 This is also the first study that documented a gradual increase initially diagnosed as a case of self-limited epilepsy with centro-
in the frequency of epileptogenic lesions with decreasing age as temporal spikes. The diagnosis on that same child would be
well as with the presence and severity of DD. For young children, changed to structural focal epilepsy if the MR imaging were to
our results are overall similar to those recently reported in a study reveal a cavernoma in the inferior frontal rolandic cortex.
that evaluated the frequency of etiologically relevant neuroimaging The strength of our study is that it evaluated a large cohort of
abnormalities in children with early-life epilepsy, most of whom consecutive children referred for new-onset unprovoked seizures
were evaluated with an epilepsy protocol brain MR imaging.20 In and who underwent a HARNESS brain MR imaging protocol.9
that study, 40% of children 3 years of age and younger were found The acquisition of MRIs was centralized, the neuroimaging stud-
to have epileptogenic lesions, with a frequency of 61% in those ies were obtained shortly after the seizure onset, and the studies
with DD compared with 24% in children with normal develop- were interpreted by an experienced neuroradiologist who was
ment.20 These results are very similar to ours, because in the group blinded to the clinical data. In addition, we were very conserva-
of children younger than 2 years of age, we identified an epilepto- tive in defining epileptogenic lesions and stratified the types and
genic lesion in 48.5%, with a 65% frequency in children with DD frequencies of lesions according to the seizure types experienced
compared with 23% in developmentally healthy children. Two by the child at the time of the initial evaluation. We elected not to
other studies that evaluated the yield of neuroimaging in children include children who underwent a non-epilepsy protocol MR
younger than 2 years of age with new-onset seizures showed simi- imaging to have a set of uniform data, especially because it was pre-
lar results, with etiologically relevant abnormalities detected in viously shown that up to 65% of studies interpreted as having nor-
42%21 and in 51%19 of children. In our study, the highest frequency mal findings would reveal a relevant lesion when a high-quality
of DD was in children younger than 2 years of age, with 61% of study was performed.24,25
children with new-onset seizures having concomitant DD. This
finding is consistent with those in previous studies that showed CONCLUSIONS
that a substantial proportion of children presenting with seizures Ideally, we believe that brain MR imaging should be performed
in early life have associated DD.22,23 in every child with new-onset unprovoked seizures, especially
Our data showed that there was a significant difference in the when sedation is not required, for several reasons: First, it would
frequencies of epileptogenic lesions according to the seizure types be in keeping with the new International League Against Epilepsy
experienced by the child at the time of evaluation. We stratified classification of the epilepsies, which emphasizes the need to con-
the seizure types into various groups based on the fact that certain sider the etiology at each step of diagnosis, including a structural
seizure types are known to occur in generalized genetic epilepsy, etiology, which should preferably be evaluated with brain MR
and others, in focal epilepsy and epileptic encephalopathies. As imaging and that will help with the syndromic classification.9,18
would be expected, the frequency of detecting an epileptogenic In addition, for children who present with a single, unprovoked
lesion was lowest in children who presented with absence and/or seizure, the presence of specific structural brain lesions could sat-
myoclonic seizures with or without generalized tonic-clonic seiz- isfy the diagnosis of epilepsy 9,17 and will also guide the need for
ures and highest in those who experienced epileptic spasms in treatment. Furthermore, the pathologic substrate identified on
clusters or frequent tonic or atonic seizures. neuroimaging can assist with the prognosis and accelerate referral
Using logistic regression, we found that only the presence of to a specialized epilepsy center.9 When brain MR imaging is not
DD (odds ratio ¼ 4.3), as well as the seizures types (odds ratio ¼ readily available, as in developing nations where the resources
2.3), remained significantly associated with the presence of an might be scarce or in case of financial constraints, it would be
epileptogenic lesion. Because the highest frequency of children useful to have guidelines to recommend when brain MR imaging
with DD was in those younger than 2 years of age, it is not sur- should be performed in children with new-onset, unprovoked
prising that the age group did not achieve statistical significance seizures. The practice parameter issued in 2000 and reaffirmed in
in the multivariate model. The results of the recursive partition- 2017 by the American Academy of Neurology, the Child
ing analysis were concordant with those of the logistic regression Neurology Society, and the American Epilepsy Society for the
and provided a tree with 6 nodes based on the seizure types and evaluation of a first nonfebrile seizure in children asserts that
the presence or absence of DD. The expected yield of detecting brain MR imaging is the preferred technique and that it should
an epileptogenic lesion in each of the various nodes could be used be seriously considered in any child with psychomotor delay,
as a decision tree to determine when brain MR imaging should focal-onset seizure, and younger than 1 year of age.7 The League
be performed. Against Epilepsy Guidelines recommend imaging (MR imaging
We purposefully avoided including the electroencephalogra- is preferred over CT when available) for infants and children
phy results in our analyses because our aim was to establish the with recent-onset epilepsy whenever localization-related epilepsy
yield of neuroimaging based on the clinical presentation alone. is known or suspected, when the epilepsy classification is in
Furthermore, because a presumed electroclinical syndrome could doubt, or when an epilepsy syndrome with a remote symptomatic
be modified by the presence of an epileptogenic lesion on the MR cause is suspected.26 However, the authors of the practice param-
imaging, including the electroencephalography results as a eter stressed that there is insufficient evidence available from the
variable can lead to circular reasoning. For example, a develop- published studies for issuing evidence-based guidelines pertain-
mentally healthy child presenting with an opercular seizure ing to neuroimaging in children with new-onset seizures.
(focal-aware seizure) and found to have rolandic maturational They also stressed the need for a large prospective study that
epileptiform discharges on the electroencephalography would be enrolls consecutive children using a standardized MR imaging
1700 Hourani Sep 2021 www.ajnr.org
acquisition protocol and that stratifies the findings according to 12. Moosa AN, Wyllie E. Focal epileptogenic lesions Handb Clin Neurol
age groups. Our study fulfills those requirements, and we hope 2013;111:493–510 CrossRef Medline
that our results will assist the relevant societies and committees in 13. Volpe JJ. Confusions in nomenclature: “periventricular leukomala-
cia” and “white matter injury”: identical, distinct, or overlapping?
formulating neuroimaging guidelines for children with new- Pediatr Neurol 2017;73:3–6 CrossRef Medline
onset seizures. 14. Humphreys P, Deonandan R, Whiting S, et al. Factors associated
with epilepsy in children with periventricular leukomalacia. J Child
Neurol 2007;22:598–605 CrossRef Medline
15. Frankenburg WK, Dodds J, Archer P, et al. Denver II Training
REFERENCES Manual. Denver Developmental Materials; 1992
1. Semah F, Picot MC, Adam C, et al. Is the underlying cause of 16. American Psychiatric Association. DSM-5 Task Force. Diagnostic
epilepsy a major prognostic factor for recurrence? Neurology and Statistical Manual of Mental Disorders (DSM-5). American
1998;51:1256–62 CrossRef Medline Psychiatric Association; 2013
2. Stephen LJ, Kwan P, Brodie MJ. Does the cause of localisation- 17. Fisher RS, Cross JH, French JA, et al. Operational classification of
related epilepsy influence the response to antiepileptic drug treat- seizure types by the International League Against Epilepsy:
ment? Epilepsia 2002;42:357–62 CrossRef Medline Position Paper of the ILAE Commission for Classification and
3. Fisher RS, Acevedo C, Arzimanoglou A, et al. ILAE official report: a Terminology. Epilepsia 2017;58:522–30 CrossRef Medline
practical clinical definition of epilepsy. Epilepsia 2014;55:475–82 18. Scheffer IE, Berkovic S, Capovilla G, et al. ILAE classification of the
CrossRef Medline epilepsies: position paper of the ILAE Commission for
4. Shinnar S, O’Dell C, Mitnick R, et al. Neuroimaging abnormalities Classification and Terminology. Epilepsia 2017;58:512–21 CrossRef
in children with an apparent first unprovoked seizure. Epilepsy Res Medline
2001;43:261–69 CrossRef Medline 19. Eltze CM, Chong WK, Cox T, et al. A population-based study of
5. Berg AT, Testa FM, Levy SR, et al. Neuroimaging in children with newly diagnosed epilepsy in infants. Epilepsia 2013;54:437–45
newly diagnosed epilepsy: a community-based study. Pediatrics CrossRef Medline
2000;106:527–32 CrossRef Medline 20. Coryell J, Gaillard WD, Shellhaas RA, et al. Neuroimaging of early
6. Kalnin AJ, Fastenau PS, Degrauw TJ, et al. Magnetic resonance life epilepsy. Pediatrics 2018;142:e20180672 CrossRef Medline
imaging findings in children with a first recognized seizure. 21. Hsieh D, Chang T, Tsuchida T, et al. New-onset afebrile seizures in
Pediatr Neurol 2008;39:404–14 CrossRef Medline infants: role of neuroimaging. Neurology 2010;74:150–56 CrossRef
7. Hirtz D, Ashwal S, Berg A, et al. Practice parameter: evaluating a Medline
first nonfebrile seizure in children: report of the quality standards 22. Battaglia D, Rando T, Deodato F, et al. Epileptic disorders with
subcommittee of the American Academy of Neurology, the Child onset in the first year of life: neurological and cognitive outcome.
Neurology Society, and the American Epilepsy Society. Neurology Eur J Paediatr Neurol 1999;3:95–103 CrossRef Medline
2000;55:616–23 CrossRef Medline 23. Altunbaş ak S , Incecik F, Hergüner Ö, et al. Prognosis of patients
8. Arabi M, Dirani M, Hourani R, et al. Frequency and stratification of with seizures occurring in the first 2 years. J Child Neurol
epileptogenic lesions in elderly with new onset seizures. Front 2007;22:307–13 CrossRef Medline
Neurol 2018;9:995 CrossRef Medline 24. Knake S, Triantafyllou C, Wald L, et al. 3T phased array MRI
9. Bernasconi A, Cendes F, Theodore WH, et al. Recommendations for improves the presurgical evaluation in focal epilepsies: a prospec-
the use of structural magnetic resonance imaging in the care of tive study. Neurology 2005;65:1026–31 CrossRef Medline
patients with epilepsy: a consensus report from the International 25. Kokkinos V, Kallifatidis A, Kapsalaki EZ, et al. Thin isotropic FLAIR
League Against Epilepsy Neuroimaging Task Force. Epilepsia MR images at 1.5 T increase the yield of focal cortical dysplasia
2019;60:1054–68 CrossRef Medline transmantle sign detection in frontal lobe epilepsy. Epilepsy Res
10. Chuang NA, Otsubo H, Chuang SH. Magnetic resonance imaging 2017;132:1–7 CrossRef Medline
in pediatric epilepsy. Top Magn Reson Imaging 2002;13:39–60 26. Gaillard WD, Chiron C, Cross JH, et al. ILAE, Committee for
CrossRef Medline Neuroimaging, Subcommittee for Pediatric. Guidelines for
11. Urbach H. Imaging of the epilepsies. Eur Radiol 2005;15:494–500 imaging infants and children with recent-onset epilepsy.
CrossRef Medline Epilepsia 2009;50:2147–53 CrossRef Medline

AJNR Am J Neuroradiol 42:1695–1701 Sep 2021 www.ajnr.org 1701