Prospective and retrospective studies that provided original data on the performance of proton interictal MRS, reported test results in detail for each individual patient, and that reported on surgical outcome based on the Engel classification or closely related criteria, were eligible for inclusion. Studies exclusively in children, patients with proven brain tumours, or patients with 'foreign tissue lesions' detected by magnetic resonance imaging (MRI) or proven by pathology, were excluded.
The studies used either chemical shift imaging or single-voxel spectroscopy, mostly at 1.5 tesla (T) although 2.0 and 4.1 T were also reported. Time of echo ranged from 50 to 272 milliseconds and time of repetition from 1.5 to 6 seconds. Chosen ratios or asymmetry indices were N-acetyl-aspartate (NAA)/choline (Cho) plus creatine-phosphocreatine (Cr), NAA/Cr, NAA/Cho, Cr/NAA and NAA alone. The reference standard was either site of surgery, Engel criteria, pathology, invasive electroencephalography (EEG), Engel criteria and MRI, or Engel criteria and invasive EEG. Classification for outcome following surgery was based on the Engel criteria, percentage reduction in seizure frequency, percentage reduction in seizures, or seizure frequency score. The duration of follow-up post surgery ranged from 3 to 78 months and the participants were aged from 3 to 66 years. All of the studies included patients with intractable temporal lobe epilepsy. Study design was either prospective or not reported.
The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.