Long-term seizure and psychosocial outcomes of vagus nerve stimulation for intractable epilepsy

Vibhangini S. Wasade, Henry Ford Health System
Lonni Schultz, Henry Ford Health System
Karthik Mohanarangan, Henry Ford Health System
Aryamaan Gaddam, Henry Ford Health System
Jason M. Schwalb, Henry Ford Health System
Marianna V. Spanaki, Henry Ford Health System

Abstract

Vagus nerve stimulation (VNS) is a widely used adjunctive treatment option for intractable epilepsy. Most studies have demonstrated short-term seizure outcomes, usually for up to 5 years, and thus far, none have reported psychosocial outcomes in adults. We aimed to assess long-term seizure and psychosocial outcomes in patients with intractable epilepsy on VNS therapy for more than 15 years. We identified patients who had VNS implantation for treatment of intractable epilepsy from 1997 to 2013 at our Comprehensive Epilepsy Program and gathered demographics including age at epilepsy onset and VNS implantation, epilepsy type, number of antiepilepsy drugs (AEDs) and seizure frequency before VNS implantation and at the last clinic visit, and the most recent stimulation parameters from electronic medical records (EMR). Phone surveys were conducted by research assistants from May to November 2014 to determine patients' current seizure frequency and psychosocial metrics, including driving, employment status, and use of antidepressants. Seizure outcomes were based on modified Engel classification (I: seizure-free/rare simple partial seizures; II: >90% seizure reduction (SR), III: 50-90% SR, IV:50% SR)=favorable outcome). A total of 207 patients underwent VNS implantation, 15 of whom were deceased at the time of the phone survey, and 40 had incomplete data for medical abstraction. Of the remaining 152, 90 (59%) were contacted and completed the survey. Of these, 51% were male, with the mean age at epilepsy onset of 9.4 years (range: birth to 60 years). There were 35 (39%) patients with extratemporal epilepsy, 19 (21%) with temporal, 18 (20%) with symptomatic generalized, 5 (6%) with idiopathic generalized, and 13 (14%) with multiple types. Final VNS settings showed 16 (18%) patients with an output current >2 mA and 14 (16%) with rapid cycling. Of the 80 patients with seizure frequency information, 16 (20%) had a modified Engel class I outcome, 14 (18%) had class II, 24 (30%) had class III, and 26 (33%) had class IV. Eighty percent said having VNS was worthwhile. Among the 90 patients, 43 patients were ≥ 18 years old without developmental delay in whom psychosocial outcomes were further analyzed. There was a decrease in the number of patients driving (31% vs 14%, p=0.052) and working (44% vs 35%, p=0.285) and an increase in the number of patients using antidepressant medication (14% vs 28%, p=0.057) at the time of survey compared to before VNS. In this subset, patients with >50% SR (60%) were taking significantly fewer AEDs at the time of survey compared to patients with unfavorable outcomes (median: 3 vs 4, p=0.045). The associations of >50% SR with the psychosocial outcomes of driving, employment, and antidepressant use were not significant, although 77% of this subset said VNS was worthwhile. This is the first study that assesses both seizure and psychosocial outcomes, and demonstrates favorable seizure outcomes of >50% SR in 68% of patients and seizure freedom in 20% of patients. A large majority of patients (80%) considered VNS therapy worthwhile regardless of epilepsy type and psychosocial outcomes.