Magnetic Resonance Imaging Of Temporal Lobe Epilepsy

Commissions of the International League Against Epilepsy: Recommendations

Who Should Have an MRI?

The Commission on Neuroimaging of the International League against Epilepsy (ILAE) (181, 182) recommends that "[i]n the non-acute situation, the ideal practice is to obtain structural neuroimaging with MRI in all patients with epilepsy, except in patients with a definite electroclinical diagnosis of idiopathic generalized epilepsy (benign myoclonic epilepsy of infancy, childhood absence epilepsy, juvenile absence epilepsy, juvenile myoclonic epilepsy), or benign epilepsy of childhood with centrotemporal spikes." Even these excluded syndromes may offer a surprise, and numerous cases where structural abnormalities are found in these electroclinical syndromes have now been reported.

What Sequences Should be Done?

The Commission on Neuroimaging of the ILAE states that "MRI is essential for presurgical evaluation. ... Epilepsy surgery should never be contemplated without an MRI examination, apart from exceptional circumstances such as a specific contraindication (e.g. cardiac pacemaker)." They recommend that a minimum of both Tl- and T2-weighted images be obtained and a three-dimensional volume acquisition with images be obtained or examined in coronal and axial orientations. It is clear that epilepsy is a specialist study in MRI and a 'routine' MRI study is not adequate for the problem of epilepsy surgery. The exact protocols will change with advances in technology as discussed in this volume (183).

Defining the Seizure Focus

The seizure focus in partial epilepsy has long been conceived of as having a brain abnormality, 'the epileptogenic lesion'; a pacemaker zone necessary for seizure generation, 'the ictal onset zone'; and a region of the brain that gives rise to the expression of the seizures, 'the symptomatogenic zone'. (These concepts are discussed in more detail in Chapter 1).

Clinical seizures, virtually by definition, demonstrate the symptomatogenic region of seizure involvement. This is the area of the brain that gives rise to clinical symptoms and is often a pointer to the area of the brain involved in seizures.

Prior to imaging techniques, EEG was the main way to tell what part of the brain was generating the seizures (Fig. 4.1). The EEG from the scalp reflects both the site of onset of seizures and its electrical spread. Although it has excellent resolution for the timing of events, EEG has poor sensitivity and poor spatial resolution. By the time electrical discharges are seen on the scalp recording, they may be a distance away from the 'seizure focus'.

Intracranial EEG is used to increase sensitivity, specificity and spatial resolution and hence is important in seizure surgery programs. It involves a major operation, and attendant risks, and still has a number of important drawbacks. It is primarily limited by the relatively small area that can be sampled. In imaging terms it has a very narrow field of view and it effectively ignores what is occurring away from the sampled area. For such a highly invasive diagnostic test it often does not provide definitive information (184). In the imaging era, intracranial studies are usually reserved for increasingly complex epilepsy.

The advent of MRI, ictal single-positron-emission computed tomography (SPECT) and positron-emission tomography (PET), and optimized imaging strategies has added direct noninvasively acquired knowledge about the lesions that are present in the brain of patients with epilepsy. These methods can also deliver information about focal brain activity as well, as is discussed in Chapter 11.

The purpose of defining the seizure focus is to remove it surgically and 'cure' the epilepsy. Increasingly, good outcome from seizures is seen to depend on complete resection of epileptogenic lesions over and above the ictal pacemaker zone, when this can be identified. The selection of surgical candidates increasingly depends on the presence of a structural abnormality and then confirmation of the epileptogenic nature of these abnormalities with functional methods such as ictal EEG and SPECT studies (185). Still, many cases of partial epilepsy appear to have no identifiable focal abnormality even in the best imaging and image interpretation centers. Part of this may result from the insensitivity of our methods, and part from the fact that the biology of seizure generation is not conveniently focal. MR methods are giving us a way to tackle these issues.

Mesial Temporal Sclerosis: Diagnosis with Magnetic Resonance Imaging

Generations of epileptologists have been fascinated with the issues that surround HS, including its origins and its causal relationship with epilepsy. Like many areas of science, the strongest opinions exist when the data is weakest. MRI has given us new insights into many of the questions that surround HS because it has given us new and compelling data. In some well studied cases, HS can be observed in relation to the onset and development of both the seizure process and the hippocampal damage. These privileged observations, in life, of patients with HS, including those not operated on, provide us with new insights into disease pathogenesis that are not possible with postsurgical or postmortem data. We therefore now reflect on a decade of observations in the MR era and bring together some of these findings into our current understanding of HS.

As we have discussed when reviewing the pathology of TLE with HS above, HS must be thought of as damage to the formed hippocampus (156). Pathologically it consists of loss of normal tissue (neuron loss and macroscopic atrophy) and gliosis and reorganization. In other words, it is a scar that appears to have occurred after the hippocampus was formed. Whether the event that caused this damage occurred in utero (making HS a 'developmental lesion') or after birth (an 'acquired lesion') is largely a semantic rather than an etio-logic argument. A similar pattern of 'damage' can be induced in a range of animal models and human disease from a variety of stimuli that can be considered to be neurotoxic.

If we accept that HS is damage to the hippocampus, longstanding questions remain: What is or are the origins of HS? What is the mechanism of damage? When does the damage occur? What is the role of other lesions? Why is it often unilateral? We believe that answers to many of these questions are taking shape.

Imaging of the Normal Hippocampus

The hippocampus is in many ways an ideal structure for MR to examine. It is well defined, it has a convenient longitudinal orientation so that cross-sectional imaging gives an excellent view of its structure even with relatively thick slices, and MRI-pathologic correlations are possible because

FIG. 4.10. A. The imaging axis typical for CT scanning is shown in this image. The location of the eye more laterally is approximated by the circle. This axis covers the brain in the minimum number of slices and does not directly expose the eye to radiation. This minimizes radiation exposure, particularly to the eye when compared to other axes. B. Parasagittal image showing the hippocampal axis. This Is approximately perpendicular to and along the long axis of the brain stem.

Continued

FIG. 4.10. Cont. C. If more lateral landmarks are used, the hippocampal axis is along the long axis of the hippocampus and perpendicular to this. Note that the orientation is similar to that shown in B. The eye is seen in this parasagittal image. D. If an image is taken in the plane shown in C, both hippocampi can be seen in the medial part of the temporal lobe. E. In this axis the coronal images cut through the hippocampus at right angles, giving the clearest possible assessment of size and internal structure. F. Parasagittal image through the hippocampus.

FIG. 4.10. Cont. C. If more lateral landmarks are used, the hippocampal axis is along the long axis of the hippocampus and perpendicular to this. Note that the orientation is similar to that shown in B. The eye is seen in this parasagittal image. D. If an image is taken in the plane shown in C, both hippocampi can be seen in the medial part of the temporal lobe. E. In this axis the coronal images cut through the hippocampus at right angles, giving the clearest possible assessment of size and internal structure. F. Parasagittal image through the hippocampus.

surgical removal is undertaken for the treatment of epilepsy. All this leads to great sensitivity and specificity in the MR assessment of the hippocampus, if the radiography of image acquisition and experience in interpretation are available.

The orientation of the hippocampus within the brain is shown in Figure 4.10. The traditional orientation of images is shown in Figure 4.10A. This axis was established in the CT era and allows coverage of the whole brain in the minimum number of slices and avoiding the radiation-sensitive lens of the eye. This orientation is familiar to most radiologists. The orientation of the 'hippocampal axis' is shown in Figure 4.10B for comparison, and roughly corresponds to the long axis of the brain stem. It is best to orient coronal images perpendicular to the long axis of the hippocampus (Fig. 4.10C) and the axial images, or reconstructions, are done perpendicular to the coronal images. The normal hippocampus in representative

FIG. 4.11. Parasagittal image showing the indentations of the dentate at the inferior border of the hippocampus. In very thin slices this can give apparent asymmetry.

BOX 4.1. The MRI Features of Hippocampal Sclerosis

How To Reduce Acne Scarring

How To Reduce Acne Scarring

Acne is a name that is famous in its own right, but for all of the wrong reasons. Most teenagers know, and dread, the very word, as it so prevalently wrecks havoc on their faces throughout their adolescent years.

Get My Free Ebook


Post a comment