Abstracts

Rationale: Current conventional therapies including adrenocorticotropic hormone and vigabatrin often fail to control infantile spasms or improve developmental outcomes, particularly in cases with structural/metabolic etiologies. Surgical resection has recently been applied for such refractory spasms, but most cases involve lesion-associated epilepsy revealed by magnetic resonance imaging (MRI). In patients without a responsible focal epileptic lesion, the surgical option of callosotomy sometimes suppresses spasms. Even in cases with residual spasms after callosotomy, asymmetric spasms and lateralization of discharges on electroencephalography (EEG) are often observed, and subsequent resective surgery for addressing those residual spasms may be possible. This study reviewed postoperative outcomes in patients with medically refractory spasms who underwent resective surgeries following callosotomy.Methods: Our pediatric epilepsy surgery program and studies were approved by the institutional review board. In total, 15 patients with refractory spasm were included in this study. All patients initially underwent total callosotomy because no causal epileptic lesions were discovered in preoperative examinations. After callosotomy, ictal semiology changed to asymmetric spasms in all cases, with additional progressive hemiparesis in three patients. Post-callosotomy workup revealed unilaterally predominant epileptic discharges on EEG and associated lateralization on single photon emission computed tomography. Finally, subsequent surgeries, including posterior quadrantectomy (n=5), frontal lobe resection (n=2), hemispherotomy (n=3) and multilobar resection/disconnection (n=5), were performed. Ranges of age at initial surgery and follow-up period were 6-85 and 3-59 months, respectively. Seizure outcome was evaluated with video-EEG at final follow-up. Developmental status (DS) before and after surgery were also measured as developmental quotient (DQ).Results: Six patients (40%) had no spasms and four patients (27%) had less frequent and clinically subtle spasms without any ictal EEG correlates on final follow-up. Series formation of spasms disappeared in all patients. Relatively good DS (DQ>50) was observed in three patients (20%) before surgical intervention and six patients (40%) at final follow-up. Preoperative DS correlated positively with postoperative DS (p<0.05).Conclusions: Infantile spasms are potentially generated by unilateral cortical activity, even if widespread/bilateral epileptogenicity is revealed on EEG. Finding the responsible focus is still difficult with pre-callosotomy EEG, neuroimaging studies and other clinical information; however, changes in ictal semiology and neurological/developmental status should be carefully observed after callosotomy, and repetitive workups using EEG and neuroimaging are required to identify the pathologic brain area/hemisphere. Finally, more radical treatment for achieving attenuation of spasms and improving development may be indicated for selected patients.