BJR
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wright, N B
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wright, N B
British Journal of Radiology 74 (2001),575-589 © 2001 The British Institute of Radiology

Review article

Imaging in epilepsy: a paediatric perspective

N B Wright, DMRD, FRCR

Royal Liverpool Children's NHS Trust, Alder Hey, Eaton Road, Liverpool L12 2AP, UK


   Abstract
Top
 Abstract
Introduction
 Definitions and concepts
 Indications for imaging
 The role of different...
 Imaging and aetiology
 Summary
 References
 
Assessment of a child with epilepsy involves a number of key stages, the most crucial being clinical evaluation where the presence of seizure activity and seizure type is identified. Subsequent imaging is not required in all children. In those selected for further investigation, imaging techniques are broadly divided into structural and functional studies. MRI currently provides the best structural data, with nuclear medicine and specialized MR techniques giving supportive functional information. CT now has a much diminished role. This review highlights the role of different imaging modalities in the investigation of childhood epilepsy, as well as some of the practicalities of imaging children, and areas where recent advances have been made. It is hoped that the overview and information provided will help both the specialist and the general radiologist make informed decisions regarding how to best image a child with epilepsy.


   Introduction
Top
 Abstract
 Introduction
Definitions and concepts
 Indications for imaging
 The role of different...
 Imaging and aetiology
 Summary
 References
 
Radiological advances have resulted in a multitude of imaging techniques that can be used to unravel the complexities of a child with a seizure disorder. However, the most crucial aspect of evaluation is not the radiological assessment, but the clinical history.

Diagnosis of epilepsy involves four key stages:

The clinical history and examination, often supplemented by video-recording by family members, form a large part of the evaluation, but further investigation with electroencephalography, neuroimaging and other non-imaging tests, e.g. blood analysis, metabolic screening and muscle biopsy, are often required to obtain a complete picture. In general, the aims of these investigations are:

and thereby

Imaging is not required in all children with epilepsy, but it plays a pivotal role in some cases. This article is intended to give an overview of its use in childhood epilepsy, highlighting the advantages of differing techniques, some of the practical difficulties encountered when imaging children, and areas where recent advances have been made. Many of the comments made here will also be applicable to imaging in adult epilepsy.


   Definitions and concepts
Top
 Abstract
 Introduction
 Definitions and concepts
Indications for imaging
 The role of different...
 Imaging and aetiology
 Summary
 References
 
An epileptic seizure is a clinical manifestation of abnormal, excessive neuronal activity arising in the grey matter of the cerebral cortex. The disorganized neuronal activity may be a purely electrophysiological event resulting in no clinically evident seizure, or it can lead to a seizure specifically related to the site of the activity, or to some secondarily activated site distant from the original source.

In general terms, epileptic disorders are separated into three broad groups: first, those in whom the brain is obviously structurally abnormal; second, those who have a predisposition to seizures owing to genetic, biochemical or microstructural changes, in whom the brain appears otherwise structurally normal; and third, those who have no structural abnormality but who develop seizures as a result of specific precipitating factors, such as hypoglycaemia or hypoxia.

Classification of seizures and epilepsy (Table 1Go)
Seizures are classified into either generalized or partial [1], with partial seizures being further divided into those without loss of consciousness (simple) and those where consciousness is lost or impaired (complex). Simple partial seizures may include motor, sensory, autonomic and psychic features. Non-motor simple partial seizures may be difficult to detect in a young child who cannot describe the symptoms. Partial seizures can progress into secondarily generalized tonic–clonic seizures, the symptoms and signs of the partial seizure forming the "aura".


View this table:
[in this window]
[in a new window]
  		Table 1. Classification of seizures

 
Once the seizure type is defined, the next step is to identify the epilepsy syndrome. This is determined by a number of factors: the seizure type, the age at onset, results of interictal and ictal electroencephalogram (EEG) findings and any additional features such as a family history or neurological findings. Identification of an epilepsy syndrome is important, as it helps to predict the prognosis, defines possible management pathways and may infer an underlying aetiology. However, it may not be possible to specify an epilepsy syndrome in 30–40% of children. There are a number of eponymous paediatric epilepsy syndromes (Table 2Go). The epilepsies and epilepsy syndromes can be classified into those that are localization-related and those that are generalized, and these can be further subdivided by aetiology. Symptomatic epilepsies are those in which the cause is known. Cryptogenic epilepsies are those in which there is a probable but unidentified cause, and idiopathic epilepsies are those in which there is no underlying cause, beyond possibly a genetic predisposition.


View this table:
[in this window]
[in a new window]
  		Table 2. A glossary of eponymous paediatric epilepsy syndromes

 

   Indications for imaging
Top
 Abstract
 Introduction
 Definitions and concepts
 Indications for imaging
The role of different...
 Imaging and aetiology
 Summary
 References
 
Recommendations for imaging in epilepsy have been drawn up by the International League Against Epilepsy Neuroimaging Commission [2]. Guidelines for specific seizure types and epilepsy syndromes are outlined in Table 3Go.


View this table:
[in this window]
[in a new window]
  		Table 3. Guidelines for specific seizure types and epilepsy syndromes

 

   The role of different imaging modalities: structure vs function
Top
 Abstract
 Introduction
 Definitions and concepts
 Indications for imaging
 The role of different...
Imaging and aetiology
 Summary
 References
 
There are two main areas in which imaging has a role to play in epilepsy; these areas are not mutually exclusive. The first is in the definition of structural abnormalities and the second is in defining areas of normal and abnormal function. The imaging techniques currently used in both areas are non-invasive. However, the use of functional imaging is usually restricted to those children being considered for surgical treatment.

MRI
MRI has revolutionized neuroimaging and provides the best method of non-invasive structural evaluation of the brain. Development of functional MRI (fMRI) also means that specific areas of brain activity can now be assessed.

Patient preparation
A number of practical issues need to be addressed when performing MRI on children. Prior to performing any investigation on a child, the need for sedation or general anaesthesia should be assessed. This is a contentious area and requires a well thought out departmental policy. MR-compatible monitoring equipment is mandatory and local guidelines need to be established and followed. A sensible amount of time, which includes considerable latitude, needs to be allocated to a child's appointment. Radiographers, nurses and ancillary staff who are familiar with the demands of paediatric practice are essential to a successful examination. Involvement of play therapists in departmental design and protocol may also be useful [3].

Technique
Routine brain imaging in paediatric practice should include a sagittal T1 weighted image of the whole brain, as many paediatric problems include midline developmental abnormalities, particularly involving the corpus callosum. T1 and T2 weighted images should be obtained, and frequently a cerebral spinal fluid (CSF) suppression (fluid attenuated inversion recovery (FLAIR)) sequence is useful, especially for identifying areas of gliosis. Although there are many possible permutations, our routine MR brain scan includes sagittal and transverse T1 weighted, transverse T2 weighted and coronal FLAIR sequences. In neonates and infants up to 6 months, it is sensible to increase the TR to 6000 ms on the T2 weighted sequences to improve the contrast between grey and non-myelinated white matter. Excellent spatial resolution and improved contrast can also be obtained using a T1 weighted inversion recovery sequence, although there may be time constraints in using this sequence.

Many children with epilepsy require further "fine tuning" of their examination to increase the likelihood of detecting an abnormality. This requires an appreciation of the clinical background and some idea as to the likely findings. When developmental structural causes seem likely, the examination should be tailored to evaluate fine anatomical detail and this generally means thin section, gradient echo volume acquisitions. Identification of cortical dysplasia is a challenging aspect of paediatric epilepsy imaging. Thin section, often coronal, volume T1 weighted sequences are essential for pinpointing the subtle cortical changes (Figure 1Go); they also provide a template for subsequent three-dimensional (3D) reconstruction of the cortical surface. The clinical and EEG findings are helpful in planning where to acquire volume information. When an active process is suspected or when there is a clinical suggestion that an event involving gliosis has occurred, techniques sensitive to detecting changes in the local chemical environment, such as FLAIR/CSF suppression, are helpful. Gliosis is an astrocytic response to tissue damage basically equivalent to brain scarring, showing on T2 weighted, proton density and FLAIR sequences as areas of high signal.



View larger version (181K):
[in this window]
[in a new window]
  		Figure 1. Coronal T1 weighted gradient echo MR volume acquisition showing bilateral areas of cortical dysplasia (arrowheads).

 
The temporal lobe
The temporal lobe is an area of special interest for radiologists dealing with childhood epilepsy, and there has been much research into MRI of this area. The lesion generally being sought is mesial temporal sclerosis, although many different processes may affect the temporal lobe. The multitude of different MR techniques that can be applied to imaging the temporal lobe reflect thedifficulty in imaging what are sometimes extremely subtle abnormalities, and also that surgical treatment can provide a "cure". The imaging techniques can be broadly divided into visual analysis ("eye-balling") and quantitative assessment of the hippocampus. When an experienced radiologist with optimized temporal lobe images performs visual analysis, a sensitivity of 93% andaspecificity of 94% can be achieved [4]. Quantitative assessment provides a relatively objective method of detecting abnormality and can involve the use of volumetric measurement and T2 relaxometry (mapping). Ipsilateral hippocampal volume loss is a reliable indicator of hippocampal sclerosis. However, there are a number of controversial issues regarding volumetric assessment. These include the considerable variation in the size of the hippocampus in the normal population [5–7], the subjective definition of the boundaries of the volume to be measured [8] and the difficulty in detecting bilateral abnormality when the emphasis is placed on the patient being their own control (relative measurement technique). Absolute measurement rather than relative assessment also has its problems. The data will need to be standardized for age, gender, hemisphere and head size, which may be difficult [9].

T2 relaxometry (mapping)
This provides an objective measure of T2 relaxation time and in the context of paediatric epilepsy generally means assessment of hippocampal grey matter. Calculation of the T2 relaxation time enables a grey scale map to be generated from the calculated mean T2 time for each pixel in the evaluated slice (Figure 2Go). In normal subjects the mean T2 time will vary depending on the MR scanner used, but in general it is less than 110–115 ms. The T2 time is elevated in mesial temporal sclerosis, and abnormal T2 relaxometry is significantly associated with intractable epilepsy [10]. One of the problems with T2 mapping in young children and infants is that the normal degree of non-myelination of the brain will result in high T2 relaxation times, making assessment difficult. Normative data on the immature brain is limited and each centre performing T2 mapping will need to define its own normal control population to make sense of the results.



View larger version (116K):
[in this window]
[in a new window]
  		Figure 2. T2 maps of the temporal lobes showing the sampling area, and histograms of the T2 times with derived average values. The left temporal lobe shows an increased T2 time (123 ms), consistent with mesial sclerosis.

 
Spectroscopy
Magnetic resonance spectroscopy (MRS) is a non-invasive method of evaluating brain metabolites. It is best accomplished by quantitative analysis, and again normative data by age are required at each institution. Usually, proton (1H) spectroscopy is performed, and can be used to measure N-acetylaspartate (NAA), choline (Cho) and creatine (Cr) in vivo. Signal acquisition can be focused on the temporal lobes or elsewhere. Both absolute and relative values can be obtained, with neuronal loss reducing NAA levels and gliosis/astrocytosis increasing Cr and Cho levels [11]. Use of the NAA/Cho+Cr signal intensity ratio has been advocated as possibly the simplest and single most useful index of spectroscopic abnormality [12]. The role of MRS is in aiding lateralization, detecting bilateral abnormalities and identifying metabolic abnormalities in epileptogenic areas. The abnormalities demonstrated by 1H MRS correlate well with interictal single photon emission CT (SPECT) [13] as well as with surgical outcome [14]. 1H MRS also has a developing role in assessing neonatal encephalopathy [15].

MRS using 31P has been used to assess energy metabolism, measuring phosphorylation potential and phopholipid metabolism in the temporal lobe interictally [16, 17]. Abnormal phosphocreatine/inorganic phosphate ratios have been found in the temporal lobes.

There has also been an interest in attempting to localize memory function using chemical shift imaging spectroscopy, and it has been suggested that it may predict post-operative outcome [18].

Functional MRI
Activated areas of the cerebral cortex show increased blood flow without an associated increase in oxygen consumption, suggesting anaerobic metabolism occurs during activation [19, 20]. This leads to an increase in the amount of oxygenated blood within the capillaries and veins in the activated area relative to non-activated areas. Deoxyhaemoglobin is paramagnetic and causes increased dephasing, which mainly affects the T2* signal, whereas oxyhaemoglobin does not have the same effect. It is this difference that forms the basis of blood oxygen level dependent (BOLD) fMRI. The increase in oxygenation level in the activated area of brain results in relative preservation of the T2* signal compared with the non-activated areas. This difference in signal intensity can be utilized to generate a function map that can be superimposed over the anatomical dataset. An increased BOLD signal has been shown in interictal states in sites of EEG discharge activity [21]. The main application in clinical practice is pre-operative localization of the motor strip, especially when seizures are related to tumours [22] and tumour-like lesions in this region [23]. There is also interest in identifying memory laterality in the temporal lobe, and early results seem encouraging, with fMRI producing complimentary information to the Wada test [24]. There may also be a role for language localization [25].

CT
The role of CT in the assessment of epilepsy has considerably diminished and is viewed as supplementary or supportive. CT can be used as an initial screening method for excluding a brain neoplasm as a cause for seizure activity, but usually MRI is more appropriate in addition to avoiding irradiation. CT is ideal for identifying the presence and extent of intracranial calcification (Figure 3Go), but this has a limited impact on patient management. CT is not as sensitive as MRI at identifying small lesions and more subtle structural changes such as those involving the temporal lobe and focal cerebral dysplasias. CT should not be the first line investigation for partial epilepsy, where MRI is the modality of choice [26, 27]. Intravenous contrast medium enhancement improves the sensitivity for vascular lesions, such as the pial angiomas in Sturge–Weber syndrome or arteriovenous malformations, and should probably be given routinely when performing head CT for epilepsy. There is also a role in the acute phase of status epilepticus, when cerebral oedema secondary to hypoxia or anoxia is suspected.



View larger version (148K):
[in this window]
[in a new window]
  		Figure 3. CT of a child with tuberous sclerosis showing periventricular calcification and low attenuation parenchymal tubers. CT remains the imaging modality of choice for identifying focal calcification.

 
Nuclear medicine
Functional assessment using radiopharmaceuticals can currently be performed using two techniques, SPECT and positron emission tomography (PET). Local availability is likely to determine which method is chosen, and their use is usually restricted to children being considered for surgery [28]. Even in such children, functional neuroimaging with SPECT or PET may be unnecessary if clinical evaluation, electroencephalography and high resolution MRI suggest unilateral mesial temporal sclerosis [29]. A common problem in paediatric practice remains the definition of a normal control group, as adult data cannot be extrapolated to the maturating brain of the infant and young child [30, 31].

SPECT utilizes radiopharmaceuticals that distribute according to regional cerebral blood flow, allowing identification of different perfusion patterns associated with epileptogenic foci both ictally and interictally. The most frequently used radiopharmaceutical is 99Tcm-labelled hexamethyl propyleneamine oxime (HMPAO) (Figure 4Go). This passes through the intact blood–brain barrier and is retained in brain tissue when converted to a hydrophilic complex dependent on the presence of glutathione [32].



View larger version (95K):
[in this window]
[in a new window]
  		Figure 4. 99Tcm-labelled hexamethyl propyleneamine oxime SPECT interictal brain images. The top two rows show a normal dataset, the bottom two rows show a number of focal defects, especially in the right frontal region.

 
Imaging is preferably performed in conjunction with EEG recording to monitor subclinical seizure activity. Injection of the radiopharmaceutical can be performed in the ictal phase (during a seizure), immediately post-ictally or interictally (at a seizure-free time) [33, 34]. The timing of the injection in relation to the onset of the seizure activity is a crucial factor [35]. In general, ictal studies are most useful, although practically are more difficult to obtain. Usually an abnormal focus shows as an increased area of activity although rarely can show as a hypoperfused area [36]. Interictal studies typically show a reduced area of activity at the site of the epileptogenic focus. Both studies are occasionally required toclarify questionable abnormalities. A meta-analysis of SPECT brain imaging reviewing 30 papers yielded sensitivities of SPECT localization relative to diagnostic evaluation of 0.97 (ictal), 0.75 (post-ictal) and 0.44 (interictal) [37]. The authors recommended interictal scans preferably combined with ictal or post-ictal images, but commented that they had insufficient data for drawing conclusions about paediatric populations or extratemporal foci. The findings of the SPECT scan must be correlated with those of any othercross-sectional imaging technique. Image co-registration, with merging of the SPECT and MR images, has been used to combine both functional and structural data. Sensitivity can also be improved by ictal–interictal subtraction images co-registered to the MRI dataset (subtraction ictal SPECT co-registered to MRI or SISCOM) [38, 39]. In this technique the interictal SPECT images are subtracted from the ictal SPECT images and the resultant functional image superimposed over the MR image to combine the functional and structural data. The practicalities of obtaining ictal SPECT studies mean that many intended ictal studies may actually be post-ictal, and therefore less sensitive. A recent study, however, has shown that the sensitivity of post-ictal studies can also be improved using the SISCOM technique [40], which may have a beneficial role in paediatric practice.

PET using 18F labelled fluorodeoxyglucose (FDG) as a reflection of the metabolic rate of glucose is also being used in the investigation of childhood epilepsy [15]. FDG uptake occurs over a 45-min period and reflects the cerebral glucose metabolic activity over that time period. This obviously makes it difficult to assess ictal states and consequently studies have usually involved interictal evaluation. In this case, areas of reduced FDG uptake are pathological (hypometabolism) and are seen in 70–80% of patients. Although co-registration with MR images can be performed, the area of abnormal activity is often too large to facilitate more accurate localization. A comparison of MRI, PET and ictal SPECT, using pathological diagnosis as a standard of reference, showed correct lateralization in 72%, 85% and 73%, respectively [41]. FDG-PET is much less successful at identifying hypometabolism in extratemporal sites, although improved high resolution techniques have been used to demonstrate frontal foci [42, 43]. Its current role is therefore primarily in lateralization of seizure activity when assessment is difficult on clinical and/or MRI grounds.

Cerebral angiography and the Wada test
Conventional angiography has no routine role in the assessment of epilepsy. It is helpful in accurately defining vascular malformations that are a source for seizure activity. It also currently has a role in pre-operative assessment when combined with neuropsychological evaluation in the Wada test. In this procedure, a barbiturate, usually sodium amytal, is administered via a catheter placed in the internal carotid artery. This anaesthetizes one cerebral hemisphere for a variable but short period. The procedure is then repeated for the other internal carotid artery. The child is given a number of neuropsychological tasks to undertake during each period of hemisphere anaesthesia in an attempt to localize the site of memory retention, since this has considerable surgical implications. In general, lateralized memory deficits using the Wada test concur with non-invasive techniques, but occasionally this is not the case and thus surgical management may be significantly altered [44]. Concerns have been raised regarding the distribution of intracarotid injections. The distribution of sodium amytal and its effect on cerebral perfusion has been investigated with a combined Wada and high resolution HMPAO SPECT injection co-registered with a patient's MRI dataset. This confirmed a variable degree of hypoperfusion of the medial temporal structures in the majority of patients undergoing the examination and suggested "partial inactivation of the memory structures is therefore a credible mechanism of action for the test" [45]. Another study confirmed that cross-perfusion only occurred into limited areas of the contralateral anterior cerebral artery, and not the contralateral temporal lobe, although the test did produce contralateral EEG changes [46].

Plain radiographs
There is no role for plain radiography in epilepsy beyond demonstrating areas of intracranial calcification, which are generally better shown by CT and the occasional craniotomy. Plain radiographs can be used beneficially in more invasive procedures, for example in identifying the position of sphenoidal and foramen ovale electrodes and intracranial EEG grids, but these are rarely performed in children.


   Imaging and aetiology
Top
 Abstract
 Introduction
 Definitions and concepts
 Indications for imaging
 The role of different...
 Imaging and aetiology
Summary
 References
 
The causes of epilepsy can be separated into genetic and acquired disorders. Some genetic causes can result in structural changes that are readily identifiable by cross-sectional imaging, but there are many epilepsy syndromes with a genetic background that have no underlying macrostructural change, such as typical absence and benign rolandic epilepsy. The indications drawn-up for imaging children with epilepsy have gone some way in trying to tease out which children do not benefit from imaging on the basis of a typical history, clinical findings and response to treatment. There is also clearly a group of patients that require imaging, such as those with a neurocutaneous syndrome or those with infantile spasms where there is a high likelihood of an underlying brain abnormality.

The causes of acquired epilepsy are numerous and include all broad groups of abnormality, such as vascular, neoplastic, degenerative, traumatic, post-infective and metabolic causes. Often there is an appropriate history, and imaging is performed to confirm the abnormality, define its extent and identify associated complications. Epilepsy may not be the main indication for imaging in these children, but may form part of the symptom complex. CT or MRI is generally performed to define the pathological process, be it structural or owing to changes in the local chemical environment, such as oedema or gliosis.

Functional studies may obviously show abnormalities in all forms of epilepsy, but these are unnecessary in the large proportion of cases where clinical and EEG features may satisfactorily localize an abnormality and where anatomical detail is paramount.

In practical terms, referrals of cases with childhood epilepsy are made to imaging departments in two main ways. One is a child with a normal examination, with little or no adverse antecedent medical history, in which the seizure type is suspicious of an underlying structural abnormality. The other is a child who has a clear neurological abnormality, with pertinent clinical findings and an adverse medical history. It is helpful to discuss imaging findings within these broad generalizations, one related to seizure type alone and the other to seizures associated with a suspected underlying cause based on an adverse history and/or examination.

Imaging specific seizure types
Temporal lobe epilepsy and mesial temporal sclerosis
Classical temporal lobe epilepsy involves a pre-seizure "aura", automatisms and repetitive motor phenomena (especially contralateral dystonic posturing of the upper limb) prior to a generalized seizure. The aura may take many manifestations, including hallucinations, abdominal features and bizarre olfactory sensations. This may be a frightening experience in young children who are unable to express themselves adequately. In suspected cases, imaging is largely tailored to identify mesial temporal sclerosis, a lesion with substantial epileptogenic potential, possibly related to febrile convulsions earlier in life. Other lesions in the temporal lobe will produce similar clinical and EEG findings. Imaging should be performed with MRI and focused on the temporal lobe. Thin section, coronal, volume T1 (gradient echo) and T2 images are essential, possibly supplemented by CSF suppression or inversion recovery sequences. In practical terms, it cannot be overemphasized that accurate positioning of the scan plane is vital for evaluation, with coronal images angled perpendicular to the hippocampus. The typical findings are of atrophy, with abnormal high signal on T2 and CSF suppression sequences in the hippocampal gyrus of the affected side (Figure 5Go). There is also loss of the normal morphology on the T1 sequences. An experienced observer with optimized images can achieve a high degree of sensitivity and specificity, up to 93% and 94%, respectively, using these features [4]. Difficulties arise when there are bilateral lesions and where the lesions are subtle. These children may require volumetric analysis, T2 mapping or spectroscopy (see earlier), with some institutes performing these studies routinely.



View larger version (170K):
[in this window]
[in a new window]
  		Figure 5. MR image of the temporal lobe showing atrophy and increased T2 signal in the left hippocampus, consistent with mesial temporal sclerosis.

 
Focus-related imaging
Clinical features from the history may suggest a focal cause for seizure activity outside the temporal lobe, occasionally supported by EEG findings. In practical terms, routine MRI is advisable with high resolution volume T1 images through the region of interest, possibly supplemented by an inversion recovery or T2 sequence. This is to identify subtle areas of cortical dysplasia (including polymicrogyria) that may be amenable to surgery. The features of cortical dysplasia are extremely variable, ranging from areas that resemble thickened, smooth cortex (pachygyria) to a cortex with a fine irregular inner and outer surface (polymicrogyria) (Figure 6Go). Some of the typical features are an abnormal gyration pattern, with similar signal intensity to normal cortex, poor differentiation of the grey–white matter interface and abnormal T2 signal from the subjacent white matter in about 20% [47]. Anomalous venous drainage is common in areas of cortical dysplasia [48]. The presence of a CSF cleft associated with a cortical dimple has recently been described as a relatively easy feature to detect, which suggests an underlying dysplasia [49]. Calcification is unusual (<5%). Occasionally, 3D cortical surface rendering (Figure 7Go) may help to evaluate the area of interest and may also assist the surgeon pre-operatively. SPECT and fMRI have also been used to assess extratemporal regions such as the occipital lobe [50, 51].



View larger version (159K):
[in this window]
[in a new window]
  		Figure 6. MR inversion recovery sequence demonstrating a subtle focal area of polymicrogyria/cortical dysplasia (arrowheads).

 


View larger version (86K):
[in this window]
[in a new window]
  		Figure 7. (a,b) Three-dimensional cortical rendering from a volume acquisition clearly showing bilateral symmetrical pachygyria of the frontal lobes.

 
Gelastic seizures
These are seizures characterized by outbursts of laughing, usually at the onset of the seizure, the laughter sounding almost mechanical in nature. They are typically associated with benign hypothalamic hamartomas. These can range from small, discrete lesions in the tuber cinereum (Figure 8Go), to large pedunculated lesions extending downwards into the pre-pontine cistern. They are commonly associated with other seizure types, such as complex partial seizures, precocious puberty and behavioural problems. MRI is the best imaging modality for identifying the lesion, which shows grey matter signal intensity on T1 and T2 weighted sequences, and fails to enhance following the administration of contrast medium. Fine, sagittal and coronal T1 weighted sections optimize visualization. Contrast medium enhancement suggests the lesion is a glioma. 1H MRS has shown abnormal NAA/creatinine ratios within the hamartomas, with normal ratios in the temporal lobes [52].



View larger version (150K):
[in this window]
[in a new window]
  		Figure 8. Coronal T1 weighted MR image through the pituitary region showing the hypothalamic hamartoma responsible for the child's gelastic seizures (arrow).

 
Aetiology-related imaging
Developmental disorders
Many children with developmental abnormalities have a high risk of epilepsy, sometimes in combination with developmental delay or focal signs. The neuronal migration disorders and other forms of cerebral dysgenesis are particularly vulnerable. The disorganized neuronal structure characterizing abnormalities such as lissencephaly and schizencephaly is a potent source of epileptogenic activity (Figure 9Go). It is therefore important to scrutinize the cerebral cortex and grey–white matter interface carefully, looking for abnormalities of the cortical thickness or pattern. Cortical dysplasia requires special mention, as its detection can be especially challenging (see above).



View larger version (188K):
[in this window]
[in a new window]
  		Figure 9. MR image showing the pial–ependymal seam of grey matter typical of schizencephaly in the right frontal lobe.

 
Children with neurocutaneous syndromes, such as tuberous sclerosis (Figure 10Go) and Sturge–Weber syndrome, are also prone to develop epilepsy. About 80% of children with tuberous sclerosis will present with infantile spasms or myoclonic seizures, with other seizure types developing with age in a large proportion. The frequency is increased if there is associated learning difficulty. Epilepsy occurs in 60–90% of children with Sturge–Weber syndrome.



View larger version (131K):
[in this window]
[in a new window]
  		Figure 10. Cerebral spinal fluid suppression (fluid attenuated inversion recovery (FLAIR)) MR image of the brain of a child with tuberous sclerosis showing multiple cortical tubers and subependymal nodules and a calcified focus in the right occipital lobe. Note the generalized increase in signal intensity of the right cerebral hemisphere related to a recent period of status epilepticus.

 
Metabolic disorders
Many metabolic abnormalities can produce seizures, although most are rare and require complex biochemical analysis for precise definition. Neurodegenerative conditions may also produce epilepsy, although this is usually in conjunction with other clinical neurological signs and dementia. Hypoglycaemia is a well known cause for epilepsy. It has been suggested that in an otherwise healthy child or adolescent presenting with seizures, the only worthwhile routine metabolic test is a blood sugar. From an imaging perspective, knowledge of the normal pattern of development of the child's brain is essential for detecting the metabolic conditions, since they often produce symmetrical abnormalities that may be subtle. Functional imaging is currently less helpful, although spectroscopy certainly has a role in evaluating the disease process [53].

Infections
Both meningitis and encephalitis can produce seizures, although the risk is small. Cerebral abscess is a particularly potent source for seizure activity. Encephalitis is more commonly associated with epilepsy, occurring in 20% of children following infection. This compares with 10% in meningitis. The risk is greater if seizures occur during the acute illness and if the child is young. ß-haemolytic streptococci and Streptococcus pneumoniae more commonly produce seizures during the acute infection and affect the younger population and are therefore more prone to producing epilepsy in the long-term. Most encephalitides produce non-specific imaging findings, but herpes simplex encephalitis can produce characteristic changes in the temporal lobes, along the olfactory grooves and involving the cingulate gyrus. Early treatment may prevent the development of complications such as epilepsy. Although CT may suggest a diagnosis of herpes encephalitis, images are often normal in the early stages [54] and MRI is more sensitive (Figure 11Go) [55].



View larger version (107K):
[in this window]
[in a new window]
  		Figure 11. (a) CT in the early stages of herpes encephalitis showing reduced attenuation in the right temporal lobe. (b) MR image showing bilateral changes involving the temporal lobes and right cingulate gyrus.

 
Congenital infections, such as the TORCH group (toxoplasmosis, rubella, cytomegalovirus and herpes simplex), may produce intractable seizures, infantile spasms and the Lennox–Gastaut syndrome (the most common intractable childhood epilepsy) (Figure 12Go). Cytomegalovirus is also prone to produce neuronal migration abnormalities.



View larger version (152K):
[in this window]
[in a new window]
  		Figure 12. MR image showing unusual, predominantly unilateral changes secondary to toxoplasmosis infection. Note the right orbital abnormality.

 
Hypoxic–ischaemic injury and cerebral palsy
Antenatal, perinatal and post-natal vascular events, either hypoxic, ischaemic or haemorrhagic, can lead to a wide range of intracranial abnormalities, some of which will be associated with epilepsy. The lesions can be unilateral or bilateral, with and without symmetry. Unilateral abnormalities may include porencephaly and focal haemorrhage. Perinatal hypoxic–ischaemic encephalopathy is a common cause of epilepsy and may result in different features depending on the severity and timing of the injury, with the pattern of abnormality giving some clue as to the underlying cause and timing of the event. This obviously can have medicolegal significance. In the acute stage of illness, ultrasound is an appropriate screening tool at the bedside. CT is a satisfactory initial investigation for identifying haemorrhage, but subsequent MRI is needed to identify the extent and pattern of damage more accurately. MRI is especially good at identifying periventricular leucomalacia, multicystic encephalomalacia and brain stem and basal ganglia changes that typify varying degrees of hypoxic–ischaemic injury (Figure 13Go).



View larger version (131K):
[in this window]
[in a new window]
  		Figure 13. MR image showing severe, widespread changes of multicystic encephalomalacia in a term child with seizures.

 
Many of the conditions that cause epilepsy may also cause cerebral palsy [56]. The incidence of epilepsy in cerebral palsy varies depending on the specific type present. Epilepsy occurs in about 20% of children with diplegic and dystonic cerebral palsy, whereas the incidence is 50–80% in the hemiplegic and quadriplegic forms.

Trauma
Seizures associated with head injury are unusual, occurring in about 5% of cases. The risk is slightly increased under the age of 5 years to between 7–9% [57]. They can be generally divided into three chronological groups: those occurring immediately, within seconds or minutes of the injury; those within the first week; and those occurring late, some months to years after the event. Certain features of the head injury increase the likelihood of late post-traumatic epilepsy [58], namely immediate or early seizures, a depressed skull fracture, intracranial haemorrhage or haematoma, focal cerebral damage (especially from penetrating injuries), prolonged post-traumatic amnesia and a genetic predisposition. There may be a clearly identified abnormality at the time of injury, such as a haematoma, or lesions may be subtler and not identified initially. A common sequence of events would include, for example, a child sustaining a closed head injury whose initial CT was normal, presenting 2–3 years later with seizures. In this setting, MRI is the imaging modality of choice and the type of lesions detectable would be gliotic scars at the grey–white matter interface (Figure 14Go), typically in the basifrontal region or temporal lobe, and shear injuries in the corpus callosum, thalamus and brain stem. The imaging technique should be guided by clinical assessment and EEG findings, but in general a FLAIR sequence is beneficial. Children who had a structural abnormality identified at the time of injury will also benefit from MRI assessment. This will define the pattern and extent of residual damage, typically encephalomalacia if the injury was severe. When the damage is extensive and surgical intervention is contemplated, SPECT imaging may be helpful to localize the epileptic focus.



View larger version (161K):
[in this window]
[in a new window]
  		Figure 14. Diffuse axonal injury. Cerebral spinal fluid suppression sequence showing a small area of parieto-occipital gliosis (arrow) following head injury in a child with post-traumatic seizures.

 
Neoplasms
Brain tumours cause 1–2% of all seizures and 4–6% of partial seizures [59]. Some of the tumours are very slow growing and imaging may initially be normal, only showing tumours months or years later. Temporal lobe tumours are often slow growing and include gliomas, dysembryoplastic neuroepithelial tumours (DNETs), oligodendrogliomas and gangliogliomas. DNETs have a high association with intractable partial complex seizures and may show characteristic features. The lesion is usually in the temporal lobe or parieto-occipital region and shows evidence of oedema with a mass lesion. There is often associated calcification and there may be scalloping of the adjacent skull vault, consistent with a chronic lesion. Follow-up imaging shows no significant change in appearance over considerable time periods. Oligodendrogliomas and low grade astrocytomas may show similar features (Figure 15Go). National protocols should be followed once a tumour is identified [60]. When the lesion lies close to the motor cortex, fMRI may be helpful in defining the precise anatomical relationship of the tumour to the motor strip (see earlier).



View larger version (170K):
[in this window]
[in a new window]
  		Figure 15. MR image showing a small enhancing nodule in the left temporal lobe in this child with intractable epilepsy. Low grade glioma.

 
Imaging at district general hospital level
What should be reasonably expected from imaging in a district general hospital? There are two main issues: ensuring the correct children are imaged and ensuring the most appropriate scans are performed in the safest manner. The indications for cross-sectional imaging have been outlined above and form the basis around which discussions between clinicians and radiologists should take place. All relevant clinical and EEG information should be available to optimize the examination, particularly if sedation or general anaesthesia is required. It may also be useful to discuss the imaging protocols with the local tertiary referral centre. Local facilities will dictate whether CT or MRI is performed, but CT may suffice if a brain tumour is suspected, although MRI should ultimately be sought. A normal CT should be viewed with caution and may give false reassurance. MRI will give the most useful information, with the focus being on structural studies rather than functional imaging. Often, high resolution volume imaging is required and this should be targeted appropriately. Complex cases, particularly where surgery is being contemplated, will inevitably be referred to tertiary centres where they may require functional studies, either with nuclear medicine or fMRI.


   Summary
Top
 Abstract
 Introduction
 Definitions and concepts
 Indications for imaging
 The role of different...
 Imaging and aetiology
 Summary
References
 
Imaging of children with epilepsy is a challenging subject and requires an understanding of the wide spectrum of pathology that affects the paediatric population. The neonate with seizures is an entirely different imaging prospect to the teenager and may have a completely different outcome. MRI will be the imaging modality of choice, often with volume acquisitions through regions of interest. More specialized MRI techniques, such as spectroscopy and functional imaging, are becoming increasingly available, but at present the results for the paediatric population should be interpreted with caution until sufficient experience and a large normal dataset are obtained. It may be that the Wada test will be replaced by fMRI in the future, but should continue to be reserved for those patients who are being considered for surgery. CT has a limited supportive role in identifying calcification within the appropriate setting. SPECT and PET certainly aid in lateralizing abnormalities where there is discordance in clinical and other imaging features, but should generally be performed at tertiary centres as part of the work-up for possible curative surgery.

Received for publication November 23, 2000. Revision received March 2, 2001. Accepted for publication April 17, 2001.


   References
Top
 Abstract
 Introduction
 Definitions and concepts
 Indications for imaging
 The role of different...
 Imaging and aetiology
 Summary
 References
 

  1. Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia 1981;22:489–501.[Medline]
  2. Commission on Neuroimaging of the International League Against Epilepsy. Recommendations for neuroimaging of patients with epilepsy. Epilepsia 1997;38:1255–6.[Medline]
  3. Pressdee D, May L, Eastman E, Grier D. The use of play therapy in the preparation of children undergoing MR imaging. Clin Radiol 1997;52:945–7.[Medline]
  4. Kuzniecky RI, Jackson GD. Temporal lobe epilepsy. In: Magnetic resonance in epilepsy (1st edn). New York: Raven Press, 1995:107–82.
  5. Watson C, Andermann F, Gloor P, et al. Anatomic basis of amygdaloid and hippocampal volume measurement by magnetic resonance imaging. Neurology 1992;42:1743–50.[Abstract/Free Full Text]
  6. Cendes F, Andermann F, Gloor P, et al. MRI volumetric measurement of amygdala and hippocampus in temporal lobe epilepsy. Neurology 1993;43:719–25.[Abstract/Free Full Text]
  7. Jack CJ, Gehring DG, Sharbrough FW, et al. Temporal lobe volume measurement from MR images: accuracy and left–right asymmetry in normal persons. J Comput Assist Tomogr 1988;12:21–9.[Medline]
  8. Jack C. MRI-based hippocampal volume measurements in epilepsy. Epilepsia 1994;35:S21–9.
  9. Watson C, Jack CR, Cendes F. Volumetric magnetic resonance imaging. Arch Neurol 1997;54:1521–31.[Abstract/Free Full Text]
  10. Jackson GD, Kim SE, Fitt GJ, Mitchell AL, Syngeniotis A, Berkovic SF. Hippocampal T2 abnormalities correlate with antecedent events and help to predict seizure intractability. Dev Neurosci 1999;21:200–6.[Medline]
  11. Gadian DG, Connelly A, Cross JH, et al. Magnetic resonance spectroscopy in children with brain disease. In: Fejerman N, Chamoles NA, editors. New trends in pediatric neurology. Amsterdam: Elsevier, 1993:23–32.
  12. Kuzniecky RI, Jackson GD. Magnetic resonance spectroscopy. In: Magnetic resonance in epilepsy (1st edn). New York: Raven Press, 1995:289–314.
  13. Cross JH, Gordon I, Connelly A, Jackson GD, Johnson CL, Neville BG, et al. Interictal 99Tc(m) HMPAO SPECT and 1H MRS in children with temporal lobe epilepsy. Epilepsia 1997;38:338–45.[Medline]
  14. Li LM, Cendes F, Antel SB, Andermann F, Serles W, Dubeau F, et al. Prognostic value of proton magnetic resonance spectroscopic imaging for surgical outcome in patients with intractable temporal lobe epilepsy and bilateral hippocampal atrophy. Ann Neurol 2000;47:195–200.[Medline]
  15. Holshouser BA, Ashwal S, Luh GY, et al. Proton MR spectroscopy after acute central nervous system injury: outcome prediction in neonates, infants and children. Radiology 1997;202:487–96.[Abstract/Free Full Text]
  16. Hugg JW, Matson GB, Duyn JH, et al. Lateralization of human focal epilepsy by 31P magnetic resonance spectroscopy imaging. Neurology 1992;42:2011–8.[Abstract/Free Full Text]
  17. Kuzniecky R, El-Gavish GA, Hetherington HP, Evanochko WT, Pohost GM. In vivo 31P nuclear magnetic resonance spectroscopy of human temporal lobe epilepsy. Neurology 1992;42:1586–90.[Abstract/Free Full Text]
  18. Pauli E, Eberhardt KW, Schafer I, Tomandl B, Huk WJ, Stefan H. Chemical shift imaging spectroscopy and memory function in temporal lobe epilepsy. Epilepsia 2000;41:282–9.[Medline]
  19. Fox PT, Raichle ME. Focal physiological uncoupling of cerebral blood flow and oxidative metabolism during somatosensory stimulation in human subjects. Proc Natl Acad Sci USA 1986;83:1140–4.[Abstract/Free Full Text]
  20. Fox PT, Raichle ME, Mintun MA, Dence C. Nonoxidative glucose consumption during focal physiologic neural activity. Science 1988;241:462–4.[Abstract/Free Full Text]
  21. Patel MR, Blum A, Pearlman JD, Yousef N, Ives JR, Saeteng S, et al. Echo-planar functional MR imaging of epilepsy with concurrent EEG monitoring. Am J Neuroradiol 1999;20:1916–9.[Abstract/Free Full Text]
  22. Connelly A, Jackson GD, Frackowiak RSJ, et al. Functional mapping of activated human primary cortex with a clinical MR imaging system. Radiology 1993;188:125–30.[Abstract/Free Full Text]
  23. Holloway V, Chong WK, Connelly A, Harkness WH, Gadian DG. Somatomotor fMRI in the pre-surgical evaluation of a case of focal epilepsy. Clin Radiol 1999;54:301–3.[Medline]
  24. Killgore WD, Glosser G, Casasanto DJ, French JA, Alsop DC, Detre JA. Functional MRI and the Wada test provide complementary information for predicting post-operative seizure control. Seizure 1999;8:450–5.[Medline]
  25. Bookheimer SY, Dapretto M, Karmarkar U. Functional MRI in children with epilepsy. Dev Neurosci 1999;21:191–9.[Medline]
  26. Cross JH, Jackson GD, Neville BGR, et al. Early detection of abnormalities in partial epilepsy using magnetic resonance. Arch Dis Child 1993;69:104–9.[Abstract/Free Full Text]
  27. Resta M, Palma M, Dicuonzo F, Spagnolo P, Specchio LM, Laneve A, et al. Imaging studies in partial epilepsy in children and adolescents. Epilepsia 1994;35:1187–93.[Medline]
  28. Robinson RO, Ferrie CD, Capra M, Maisey MN. Positron emission tomography and the central nervous system. Arch Dis Child 1999;81:263–70.[Free Full Text]
  29. Polkey CE. Surgical treatment of epilepsy in children. In: Wallace S, editor. Epilepsy in children. London: Chapman and Hall, 1996:561–600.
  30. Chugani HT, Phelps ME. Maturational changes in cerebral function in infants determined by 18FDG positron emission tomography. Science 1986;231:840–3.[Abstract/Free Full Text]
  31. Kinnala A, Suhonen-Polvi H, Kero P, et al. Cerebral metabolic rate for glucose during the first six months of life: an FDG positron emission study. Arch Dis Child 1996;74:F153–7.
  32. Neirinckx RD, Burke JF, Harrison RC, Forster AM, Andersen AR, Lassen NA. The retention mechanism of technetium-99m HMPAO: intracellular reaction with glutathione. J Cereb Blood Flow Metab 1988;8(Suppl.):S4.[Medline]
  33. Otsubo H, Hwang PA, Gilday DL, Hoffman HJ. Location of epiletic foci on interictal and immediate postictal single photon emission tomography in children with localisation-related epilepsy. J Child Neurol 1995;10:375–81.[Abstract/Free Full Text]
  34. O'Brien TJ, Zupanc ML, Mullan BP, O'Connor MK, Brinkmann BH, Cicora KM, et al. The practical utility of performing peri-ictal SPECT in the evaluation of children with partial epilepsy. Pediatr Neurol 1998;19:15–22.[Medline]
  35. Cross JH, Gordon I, Jackson GD, Boyd SG, Todd-Pokropek A, Anderson PJ, et al. Children withintractible epilepsy: ictal and interictal 99TcM HMPAO single photon emission computed tomography. Dev Med Child Neurol 1995;37:673–81.[Medline]
  36. Duncan R, Patterson J, Roberts R, Hadley DM, Bone I. Ictal/postictal SPECT in the presurgical localisation of complex partial seizures. J Neurol Neurosurg Psychiatry 1993;56:141–8.[Abstract/Free Full Text]
  37. Devous MD Sr, Thisted RA, Morgan GF, Leroy RF, Rowe CC. SPECT brain imaging in epilepsy: a meta-analysis. J Nucl Med 1998;39:285–93.[Abstract/Free Full Text]
  38. O'Brien TJ, So EL, Mullen BP, Hauser MF, Brinkmann BH, Bohnen NI, et al. Subtraction ictal SPECT co-registered to MRI improves clinicalusefulness of SPECT in localizing thesurgical seizure focus. Neurology 1998;50:445–54.[Abstract]
  39. Vera P, Kaminska A, Cieuta C, Hollo A, Stievenart JL, Gardin I, et al. Use of subtraction ictal SPECT co-registered to MRI for optimising the localisation of seizure foci in children. J Nucl Med 1999;40:786–92.[Abstract/Free Full Text]
  40. O'Brien TJ, So EL, Mullan BP, Hauser MF, Brinkmann BH, Jack CR, et al. Subtraction SPECT co-registered to MRI improves postictal SPECT localization of seizure foci. Neurology 1999;52:137–46.[Abstract/Free Full Text]
  41. Won HJ, Chang KH, Cheon JE, Kim HD, Lee DS, Han MH, et al. Comparison of MR imaging with PET and ictal SPECT in 118 patients with intractable epilepsy. Am J Neuroradiol 1999;20:593–9.[Abstract/Free Full Text]
  42. Swartz BE, Khonsari A, Brown C, Mandelkern NM, Simkins F, Krisdakumtorn T. Improved sensitivity of 18FDG-positron emission tomography scans in frontal and "frontal plus" epilepsy. Epilepsia 1995;36:388–95.[Medline]
  43. da Silva EA, Chugani DC, Muzik O, Chugani HT. Identification of frontal lobe epileptic foci in children using positron emission tomography. Epilepsia 1997;38:1198–208.[Medline]
  44. Spencer DC, Morrell MJ, Risinger MW. The role of the intracarotid amobarbital procedure in evaluation of patients for epilepsy surgery. Epilepsia 2000;41:320–5.[Medline]
  45. de Silva R, Duncan R, Patterson J, Gillham R, Hadley. Regional cerebral perfusion and amytal distribution during the Wada test. J Nucl Med 1999;40:747–52.[Abstract/Free Full Text]
  46. Hong SB, Kim KW, Seo DW, Kim SE, Na DG, Byun HS. Contralateral EEG slowing and amobarbital distribution in the Wada test: an intracarotid SPECT study. Epilepsia 2000;41:207–12.[Medline]
  47. Barkovich AJ, Kjos B. Non-lissencephalic cortical dysplasia: correlation of imaging findings with clinical deficits. Am J Neuroradiol 1992;13:95–103.
  48. Barkovich AJ. Abnormal vascular drainage inanomalies of neuronal migration. Am J Neuroradiol 1988;9:939–42.
  49. Bronen RA, Spencer DD, Fulbright RK. Cerebrospinal fluid cleft with cortical dimple: MR imaging marker for focal cortical dysgenesis. Radiology 2000;214:657–63.[Abstract/Free Full Text]
  50. Duncan R, Biraben A, Patterson J, Hadley D, Bernard AM, Lecloirec J, et al. Ictal single photon emission computed tomography in occipital lobe seizures. Epilepsia 1997;38:839–43.[Medline]
  51. Masuoka LK, Anderson AW, Gore JC, McCarthy G, Spencer DD, Novotny EJ. Functional magnetic resonance imaging identifies abnormal visual cortical function in patients with occipital lobe epilepsy. Epilepsia 1999;40:1248–53.[Medline]
  52. Tasch E, Cendes F, Li LM, Dubeau F, Montes J, Rosenblatt B, et al. Hypothalamic hamartomas and gelastic epilepsy: a spectroscopic study. Neurology 1998;51:1046–50.[Abstract/Free Full Text]
  53. Wang ZJ, Zimmerman RA. Proton MR spectroscopy of pediatric brain metabolic disorders. Neuroimaging Clin N Am 1998;8:781–807.[Medline]
  54. Cameron PD, Wallace SJ, Munro J. Herpes simplex virus encephalitis: problems in diagnosis. Develop Med Child Neurol 1992;34:134–40.[Medline]
  55. Bale JF, Andersen RD, Grose C. Magnetic resonance imaging of the brain in childhood herpes virus infections. Pediatr Infect Dis J 1987;6:644–7.[Medline]
  56. Aksu F. Nature and prognosis of seizures in patients with cerebral palsy. Develop Med Child Neurol 1990;32:661–8.[Medline]
  57. Annegers JF, Grabow JD, Groover RV, Laws ER, Elveback LR, Kurland LT. Seizures after head trauma: a population study. Neurology 1980;30:683–9.[Abstract/Free Full Text]
  58. Kieslich M, Jacoby G. Incidence and risk factors of post-traumatic epilepsy in childhood. Lancet 1995;345:187.
  59. Aicardi J. Epilepsies as a presenting manifestation of brain tumours and other selected brain disorders. In: Epilepsy in children (2nd edn). New York: Raven Press, 1994:334–53.
  60. Griffiths PD. A protocol for imaging paediatric brain tumours. United Kingdom Children's Cancer Study Group and Societe Francaise d'Oncologie Pediatrique Panelists. Clin Radiol 1999;54:558–62.[Medline]



This article has been cited by other articles:


Home page
 ImagingHome page
H J Williams
Imaging the child with developmental delay
Imaging, December 1, 2004; 16(2): 174 - 185.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wright, N B
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wright, N B

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
BJR DMFR IMAGING  ALL BIR JOURNALS