Abstracts

Rationale: Vagus nerve stimulation (VNS) is a treatment option for patients with refractory epilepsy who are not candidates for surgical resection. Most outcome analysis of VNS has used the percentage of patients with a 50% reduction in seizure frequency. Few VNS trials discuss the percentage of patients who became seizure-free, or compare their results to those of resective surgery. We decided to compare VNS with brain surgery by using the Engel outcome classification.Methods: The medical records of all patients who have been treated with VNS at Dartmouth-Hitchcock Medical Center and Ghent University Hospital with a follow-up of at least six months were evaluated. The patient population consisted of 116 patients with a mean follow-up of 2.5 years. 51.7% were male and 48.3% female, with a mean age of 33 years. The effect of VNS treatment on seizure control was assessed using the Engel classification, placing all patient outcomes in one of four categories (class I-IV). In patients with mental retardation, caregivers’ assessment of quality of life was taken into account in deciding between class III and class IV outcomes. Sub-group analysis looked at outcome in patients with: 1: clusters of seizures, 2: rapid cycling stimulation, 3: mild or severe mental retardation, 4: early or late age at start of VNS, 5: specific risk factors, 6: specific seizure types.Results: Among all patients (N=116), 4.3% had a class I outcome (seizure-free), 18.1% achieved a class II outcome (rare seizures), 36.2% had a class III outcome (worthwhile improvement), and 41.4% of patients were in class IV (no improvement). When the patient population was sub-grouped to look at particular patient characteristics, classes I and II, and classes III and IV were pooled. This was done to simplify statistical analysis of small numbers. VNS success for each sub-group was defined as the percentage of patients that achieved a class I or class II outcome. Compared to the success rate for the entire group(4.3+18.1=22.4%), the various sub-groups listed above were similar. One difference was that patients with complex partial seizures (N=54) responded better, and had a combined class I and II outcome of 33.3% (p=0.0664). Also, trends were noted, with only 4.8% of the severely retarded (N=21), and 14.7 % of the mildly retarded (N=34) having class I or class II seizure outcomes.Conclusions: Following resective surgery, in most series 50-75% of patients are seizure-free and combined class I and II outcomes are about 80%. In our series, results from VNS are clearly inferior to surgery but comparison is possible. With combined class I and II outcomes of 20-30%, VNS is a viable alternative when resective surgery is not feasible. Mentally intact patients with complex partial seizures do best in terms of seizure control. However, among the mentally retarded, many of the class III outcomes were assigned on the basis of quality-of-life improvements, and those may justify use of VNS. Sources of Funding: This research was supported by Departmental funds and in part by a research grant from Cyberonics.