Response to antiepileptic drugs in partial epilepsy with structural lesions on MRI

Although partial epilepsy with structural lesions on MRI (lesional epilepsy) is less favorably responsive to antiepileptics than those without structural lesions on MRI, the response to antiepileptics in lesional epilepsy is a heterogeneous condition. There is growing evidence that the extent of epileptic network beyond the visible lesion on MRI may be related with the response to antiepileptics. The aims of this study are to clarify whether (1) the epilepsy network beyond the visible lesion on MRI, or (2) duration of lesional epilepsy on MRI are related with the response to antiepileptics or not.

The inclusion criteria for this study were (1) having structural lesions on MRI, (2) taking antiepileptics for at least 1 year, and (3) age ≥13 years old. The definition for drug-resistance epilepsy was a failure of adequate trials of two tolerated, appropriately chosen and used antiepileptics to achieve sustained seizure freedom. The duration was defined as the interval between the start of antiepileptics and the last follow-up. We defined the lesion-plus group as the structural lesions on MRI that has wider spread of epileptic network beyond the visible lesions on MRI, such as hippocampal sclerosis and malformation of cortical development. Lesion-restriction group was defined as the epileptic network being believed to be limited on the structural lesions.

We found 234 patients with lesional epilepsy, who met the inclusion criteria. Of these 234 patients, 115 patients were male (49%) and 119 patients were female (51%). The median age was 22 years old (range 13-78 years old) and the median duration was 131 months (range 12-516 months). Forty percent (90/234 patients) were intractable to antiepileptics. Of the structural lesions on MRI, hippocampal sclerosis was most frequent (N=90). Other structural lesions were malformation of cortical development (N=38), cerebromalatic lesions related with trauma (N=34), tumor (N=19), cystic lesion (N=15), cerebral infarction (N=11), vascular malformation (N=10), and other miscellaneous lesion (N=24). Lesion-plus group had significantly higher drug-resistance epilepsy than cystic lesions on MRI (60/128 vs. 2/15, p=0.013 by Fisher's exact test). There was a tendency of having more drug-resistance epilepsy in the lesion-plus group than the lesion-restriction group (56/121 vs. 30/89, p=0.09 by Chi-square test). The median duration in drug-resistance epilepsy was significantly longer than that of medically controlled epilepsy (178 months (range 23-516 months) vs 102 months (range 12-479 months), p<0.0001 by Mann-Whitney test). In addition, duration was only the significant variable associated with drug-resistance epilepsy in lesional epilepsy by multiple logistic regression analysis (p=0.02 for overall model fit).

In lesional epilepsy, hippocampal sclerosis and malformation of cortical development are more intractable to antiepileptics, reflecting wider epileptic network beyond the visible lesion. In addition, the response to antiepileptics may be expected to decrease when the duration is prolonged.