For H-1 magnetic resonance spectroscopy (MRS) studies in epilepsy in v
ivo, the main signals of interest have been those from N-acetyl aspart
ate (NAA), creatine + phosphocreatine (Cr), choline-containing compoun
ds (Cho), and lactate (Lac). Several lines of evidence have indicated
that NAA is located primarily within neurons. A reduction of the NAA s
ignal is frequently interpreted, therefore, as reflecting loss or dysf
unction of neurons. Such a reduction in NAA has been found in practice
in many cases where neuronal loss would be expected clinically, and s
uch abnormalities have been shown to be present in patients with intra
ctable focal epilepsy. The acquisition of spectroscopic data in vivo i
s of limited use unless the spatial origin of the signals obtained is
known. It is necessary, therefore, to limit the extent of the volume f
rom which signal is acquired and to relate this volume to the anatomy
of the region under investigation. For this reason, number of common s
patial localisation techniques are outlined, and their consequences de
scribed. In vivo MRS studies in epilepsy to date have been directed la
rgely at the temporal lobes of patients with intractable epilepsy in w
hom surgical treatment is under consideration. Results are presented f
rom a number of institutions which demonstrate that H-1 MRS can contri
bute to the lateralization of the epileptic focus and to the identific
ation of bilateral pathology. In providing information both on static
metabolic abnormalities, primarily through the NAA signal, and on dyna
mic changes associated with seizure activity via the lactate signal, M
RS offers the possibility of obtaining information on cerebral metabol
ism that is likely to have significant consequences for the management
of patients with epilepsy.