The evidence from the literature suggests that VNS is a safe and effective method of treatment when added to the existing regimen of patients with refractory epilepsy. However, the authors do not feel able to recommend its use. They point out that although the evidence for VNS in epilepsy comes from RCTs of acceptable quality in other respects, the outcome measures are impossible to interpret in terms of quality of life. Questions also remain on the mode of action of VNS, which patients are likely to respond, QoL, cognitive function and cost benefit, and impact on carers.