Voxel-based comparison of preoperative FDG-PET between mesial temporal lobe epilepsy patients with and without postoperative seizure-free outcomes
- First Online:
- Cite this article as:
- Takahashi, M., Soma, T., Kawai, K. et al. Ann Nucl Med (2012) 26: 698. doi:10.1007/s12149-012-0629-9
This study aims to elucidate differences in preoperative cerebral glucose metabolism between patients who did and did not become seizure free after unilateral mesial temporal lobe epilepsy (mTLE) surgery. We hypothesized that regional glucose metabolism on preoperative fluorodeoxyglucose positron emission tomography (FDG-PET) in patients with seizure-free outcomes differed from that in patients who were not seizure free after appropriate epilepsy surgery for unilateral mTLE. In this study, we compared preoperative FDG-PET findings between these two patient groups by applying a statistical analysis approach, with a voxel-based Asymmetry index (AI), to improve sensitivity for the detection of hypometabolism.
FDG-PET scans of 28 patients with medically refractory mTLE, of whom 17 achieved a seizure-free outcome (Engel class 1 a–b) during a postoperative follow-up period of at least 2 years, were analyzed retrospectively. Voxel values were adjusted by the AI method as well as the global normalization (GN) method. Two types of statistical analysis were performed. One was a voxel severity analysis with comparison of voxel values at the same coordinate, and the other was extent analysis with comparison of the number of significant voxels in the anatomical areas predefined with Talairach’s atlas.
In the voxel severity analysis, significant hypometabolism restricted to the ipsilateral temporal tip and hippocampal area was detected in the postoperative seizure-free outcome group as compared to controls. No significant area was detected in the non-seizure-free group as compared to controls (family-wise error corrected, p < 0.05). With extent analysis using a low threshold, the extents of hypometabolism in the ipsilateral hippocampal, frontal and thalamic areas were larger in the seizure-free than in the non-seizure-free group (p = 0.01, 0.03 and 0.01, respectively.) On the other hand, in the contralateral frontal and thalamic areas, extents of hypometabolism were smaller in the seizure-free than in the non-seizure-free group (p = 0.01 and 0.01).
We found the preoperative distribution of hypometabolism to differ between the two patient groups. Severe hypometabolism restricted to the unilateral temporal lobe, with ipsilateral dominant hypometabolism including mild decreases, may support the existence of an epileptogenic focus in the unilateral temporal lobe and a favorable seizure outcome after mTLE surgery.