Long-term outcome of vagus nerve stimulation for drug-resistant epilepsy using continuous assessment, with a note on mortality.

PURPOSE:To examine the outcome of vagus nerve stimulation (VNS) for drug-resistant epilepsy using data from a National Health Service VNS clinic. METHODS:Clinical records of patients implanted with VNS for epilepsy between1995 and 2010 were examined. Patients were selected for study who had at least one year of therapeutic stimulation (minimum 1 mA stimulator current) and follow-up by our service with analysable electronic records, providing continuous assessment of seizure control during available follow-up. Seizure status at each attendance was assessed and graded 1-4 (1=seizure free or <5 seizures/year; 2 =≥50%reduction in seizure frequency; 3=<50% reduction; 4=no improvement compared to baseline). Responders were those whose grades improved consistently (Grades 1,2 and 3). RESULTS:Of 464 patients, 171 fulfilled the inclusion criteria and were divided into three groups: a) Responders (n = 81); b) non-responders (n = 80) and c) others (n = 10), the latter showing a late step-wise change (six improved; four deteriorated). After initial ramping up of current, groups were very stable over subsequent periods varying from one to 12 years (median 3.8 years). Sixteen patients died, 10 of non-epilepsy causes with 6 epilepsy-related deaths. There was a significant relation between epilepsy-related deaths and response (p < 0.00001). Patients with longer time as non-responders had more likelihood of suffering an epilepsy death than responders, though numbers were small. CONCLUSION:This study shows that meaningful data can be obtained retrospectively from routine clinic records. In this cohort about half of patients treated with VNS responded and the response generally remaining stable over time.