Object. The goal of this study was to define the incidence and clinical significance of amygdala sclerosis (AS) in patients with temporal lobe epilepsy (TLE).
Methods. Surgical specimens of the lateral amygdaloid nucleus and the hippocampus excised from 71 patients who were treated for medically intractable TLE were quantitatively evaluated using a computer-assisted image-analysis system and compared with 10 normal autopsy specimens. Densities of neurons and reactive astrocytes in the patients with TLE were correlated with clinical, neuropsychological, and depth-electroencephalography data. The neuron counts of the lateral amygdaloid nucleus did not correlate with various presumed etiological factors of TLE including hereditary seizures, birth complications, febrile convulsions, traumatic brain injury, infections, seizure semiology, and epileptological outcome. However, patient age at surgery was significantly higher (mean difference 10 years) when AS was present, as compared with patients without AS (p < 0.01). Seizure origin, as determined by using amygdalohippocampal depth electrodes, did not correlate with the presence or absence of AS. Neuropsychologically, there was a significant correlation between the neuronal densities of the lateral amygdaloid nucleus and both preoperative visual recognition and postoperative deterioration of short-term verbal memory performance (p < 0.05).
Conclusions. Except for the relatively long history of epilepsy, the presence of AS is not associated with specific clinical or electrocorticographic features of mesial TLE. However, patients without AS are particularly at risk for deterioration of short-term verbal memory following amygdalohippocampectomy.