Long-term outcome after callosotomy or vagus nerve stimulation in consecutive prospective cohorts of children with Lennox–Gastaut or Lennox-like syndrome and non-specific MRI findings

PURPOSE There is currently no resective (potentially curative) surgical option that is useful in patients with Lennox-Gastaut syndrome. Palliative procedures such as callosotomy (Cx), vagus nerve stimulation (VNS) or deep brain stimulation have been offered. We compared the outcomes after Cx or VNS in two consecutive prospective cohorts of patients with generalised epilepsy. METHODS Twenty-four patients underwent callosotomy from 2006 to 2007 (Group 1); 20 additional patients were submitted to VNS from 2008 to 2009 (Group 2). They had generalised epilepsy of the Lennox-Gastaut or Lennox-like type. They were submitted to a neurological interview and examination, interictal and ictal video-EEG, high resolution 1.5T MRI, and cognitive and quality of life evaluations. The two-year post-operative follow-up results were evaluated for each patient. RESULTS The final mean stimuli intensity was 3.0 mA in the Group 2 patients. Seizure-free patients accounted for 10% in Group 1 and none in Group 2. Ten and sixteen percent of the Group 1 and 2 patients, respectively, were non-responders. Improvements in attention and quality of life were noted in 85% of both Group 1 and 2 patients. Rupture of the secondary bilateral synchrony was noted in 85% of Group 1 patients; there was no EEG modification after VNS in Group 2. Both procedures were effective regarding the control of atypical absences and generalised tonic-clonic seizures. Both procedures were not effective in controlling tonic seizures. Callosotomy was very effective in reducing the frequency of atonic seizures, but VNS was ineffective. In contrast, callosotomy was not effective in reducing myoclonic seizures, whereas VNS was. DISCUSSION Callosotomy might be preferred as the primary treatment in children with Lennox-Gastaut syndrome, and no specific findings on MRI if atonic seizures prevail in the patient's clinical picture; when myoclonic seizures prevail, the same might hold true in favour of VNS. When atypical absence or generalised tonic-clonic seizures are the main concern, although both procedures carry similar effectiveness, VNS might be considered a good option as an initial approach, taking into account the adverse event profile. Patients should be advised that both procedures are not very effective in the treatment of tonic seizures.