Magnetic Resonance Spectroscopy in Epilepsy: Technical Issues

Over the last 5 years, there has been a dramatic expansion in the use of magnetic resonance (MR) spectroscopic measures for the evaluation of metabolic abnormalities associated with epilepsy. Single-voxel and spectroscopic imaging techniques have both been successfully used in a number of ways, for example, to aid presurgical lateralization of the seizure focus in temporal lobe epilepsy (TLE), to identify bilateral abnormalities (which can be present even in cases where the seizures are well lateralized), and to demonstrate recovery of the metabolic status of unresected tissue after successful surgery. Most studies have used the H nucleus, although P MR spectroscopy (MRS) has also demonstrated its ability to demonstrate impairments in bioenergetics. H MRS investigations have used a variety of different types of acquisition conditions; many groups have shown how relatively simple single-voxel, long echo-time (TE 4 135 or even 270 ms) acquisition can provide important information through the three main signals from N-acetyl aspartate (NAA), creatine + phosphocreatine (Cr), and choline-containing compounds (Cho), whereas other groups have concentrated on the development and application of more sophisticated spectroscopic techniques, which can provide more detailed information about the spatial distribution of metabolites and can provide access to a wider range of compounds, including glutamate/ glutamine. We outline in brief some of the recent developments in these techniques, which benefit from the higher field strengths that are now becoming more routinely available. We also briefly describe some recent applications of MRS to cortical dysplasia–induced epilepsy. One of the main questions that should be addressed is the extent to which MRS techniques, either the simple or the more complex ones, provide information that is additive to optimized MRI. In response to this question, it is apparent that the reduction in NAA that is often seen in patients with epilepsy, and which is attributed to neuronal loss or dysfunction, reflects damage that is frequently more widespread than the damage that is seen on MRI. Indeed, MRS may sometimes reveal abnormalities when no abnormalities at all are seen on MRI. For example, one study reported on seven patients with intractable temporal lobe epilepsy, who had normal routine MRI, normal hippocampal volumes, and normal hippocampal and amygdala quantitative T 2 values. Singlevoxel H MRS of the medial temporal lobes revealed abnormal NAA/(Cho + Cr) ratios in five of these seven MRI-negative patients, indicating the presence of widespread or diffuse disease that was not visible on extensive MRI investigation (1). More generally, it is clear that the type of information that MRS can provide about cellular biochemistry and pathophysiology, including, for example, the detection of a focal lactate signal in the early postictal phase, complements the information that is available from the array of other techniques that are used in the evaluation of patients with epilepsy.