Abstract
Article abstract We treated 24 generalized epilepsy patients with vagus nerve stimulation (VNS), comparing seizure rates during a 1-month baseline with 3 months of VNS. Median seizure rate reduction was −46%. Sixteen of the 24 patients had better than a −30% reduction and 11 of the 24 patients had better than a −50% reduction in seizure rate. A mild cough during stimulation occurred in six patients. Patients with higher baseline seizure rates and later ages at epilepsy onset had the best responses to VNS. Our findings suggest VNS is an effective treatment for medication-resistant generalized epilepsy even in patients as young as 4 years.
Vagus nerve stimulation (VNS) has efficacy for localization-related epilepsy.1,2 Laboratory studies have shown that VNS is effective in partial-onset and primary generalized seizures, including maximal electric shock and intraperitoneal pentylenetetrazol (PTZ) in rats3,4 and IV PTZ and strychnine in dogs.5 Several pilot studies suggested that VNS benefited patients with idiopathic or symptomatic generalized epilepsy.6-9 We now describe VNS treatment of 24 generalized epilepsy patients.
Methods.
Patient enrollment.
This analysis derives from a multicenter, prospective, open label trial (Cyberonics E04 [see Appendix]) of adjunctive VNS for medication-resistant seizures. For that trial, patients were required to 1) have ≥ one seizure per month, 2) be older than 3 years, and 3) have no cardiac or progressive neurologic disease. The study was approved by the institutional review boards, and informed consent was obtained. This report describes the responses of all E04 patients with only generalized seizures and only generalized epileptiform activity (23) or generalized slowing (1) on EEG.
Design of the study.
After a 1-month baseline, we implanted the VNS system; stimulation parameters replicated the “high stimulation” groups previously described.1,2 We did not change antiepileptic drugs (AEDs).
Clinical observations.
We reviewed the patients’ neuroimaging, EEG, cause of epilepsy, type of seizure, and neurologic examination data. We classified epilepsy as idiopathic or symptomatic. The daily number of each type of seizure was recorded. We collected diaries, recorded vital signs, and queried patients concerning adverse events (AEs) monthly.
Statistical analysis.
Our dependent variable was percent change in seizure rate in the first 3 months of VNS: 
When normality of the data could be demonstrated, we used parametric statistics. Repeated measures on single samples were compared with paired t-tests. Between-group comparisons were made with independent t-tests. When normality could not be demonstrated, we used nonparametric statistics. Repeated measures on single samples were compared with the Wilcoxon matched-pairs sign-rank test. Between-group comparisons were made with the Mann-Whitney U test. All tests were two-tailed. We considered p values < 0.05 significant.
We performed a multiple regression analysis (SPSS, Chicago, IL). The dependent variable was percent change in seizure rate. The five independent variables were baseline seizure rate/month, age at epilepsy onset, number of AEDs coadministered, number of seizure types, and epilepsy class. Log transformation of the percent change in seizure rate was performed when the data were not normally distributed.
Results.
Patient characteristics.
Twenty-five patients had VNS implants (table 1). For one patient, caregivers were unable to maintain reliable seizure diaries (a 5-year-old girl with symptomatic epilepsy). Thirteen patients were male and 11 were female. Seven patients had idiopathic and 17 patients had symptomatic epilepsy (10 cryptogenic, 4 postinfectious, 3 congenital brain injury). Eleven patients had multiple seizure types.
Patient characteristics
Changes in seizure rate with VNS.
For all seizure types in all patients, VNS produced a median seizure rate change of −46% (range −85% to +130%) (p = 0.004). Sixteen of 24 patients had better than a −30% reduction, and 11 of 24 patients had better than a ∼50% reduction. Generalized tonic-clonic (GTC) seizures had a median change of −41% (range −100% to +350%) (p = 0.029). One 12-year-old with symptomatic epilepsy and one 10-year-old with idiopathic epilepsy had increased seizures (+130% and +60%).
Among those with symptomatic epilepsy, VNS produced a median seizure rate change of −40% (range −85% to +130%) (p = 0.037); among those with idiopathic epilepsy, VNS produced a median change of −60% (range −84% to +60%) (p = 0.111). This difference between epilepsies was not significant (p = 0.332).
VNS reduced GTC seizures in patients with symptomatic epilepsy by −33% (range −100% to +350%) (p = 0.048) (table 2). The numbers of patients in each of the other seizure and epilepsy groups were small (3, 4, or 8), and the seizure rate reductions were not significant. No patient became seizure-free. Six patients became free of one seizure type. In five out of these six, the baseline rate of that seizure type was < four per month.
Percent change relative to baseline in rate of various types of seizures
Multiple regression analysis showed that a higher baseline seizure rate predicted a better reduction in seizure rate (p < 0.0001). Baseline seizure rate accounted for 49.6% of the variability. Furthermore, an older age at epilepsy onset also predicted VNS responsiveness (p < 0.01). Age at epilepsy onset accounted for 15.2% of the variability. Number of AEDs coadministered, number of seizure types, and epilepsy class were not significant predictors of responsiveness.
Adverse events.
All AEs were “mild” except one “moderate” cough and one “moderate” anorexia (table 3). One patient had an incisional infection, treated with antibiotics and surgical debridement. The median heart rate at 3 months (79 beats per minute [BPM], range 70 to 119) was lower than at baseline (88 BPM, range 58 to 125) (p = 0.0293). At the same visits, there were no differences in mean arterial blood pressures (p = 0.707).
Adverse events on vagus nerve stimulation
Discussion.
Efficacy of VNS in generalized epilepsy. Our efficacy variable was the change in seizure rate during 3 months of VNS compared with a 1-month baseline. This 1 month may be shorter than optimal, as seizure rates vary greatly. However, we do not believe our results are due to regression to the mean. Twenty-two of 24 patients improved. We do not believe our results are due to a placebo effect. In two previous double-blind, randomized, controlled trials, the “low” stimulation (placebo) patients had seizure rate reductions of −6.1%1and −15.2%.2 In contrast, our median seizure rate reduction was −46%.
Idiopathic epilepsy patients improved more than those with symptomatic epilepsy (−60% versus −40%). However, there was considerable overlap in the responses of these groups, the number in each group was small, and the difference between groups was not significant.
Among symptomatic patients, generalized tonic seizures responded better than GTC seizures (−70% versus −33%), which is similar to a trend noted in the response to corpus callosum section in this type of patient. This raises the possibility that VNS may disrupt interhemispheric spread and synchronization of epileptogenic activity, much as corpus callosum section does.
The only predictors associated with a better response were epilepsy onset at a later age and a higher baseline seizure rate. Our median age at epilepsy onset was only 2 years. Thus, an “early” age of epilepsy onset implies presentation in the infantile period. Epilepsy beginning in these early years may be resistant to VNS therapy, whereas seizure disorders developing later may be more responsive to VNS. Perhaps a certain plasticity in central vagus-activated pathways is required for VNS antiepileptic responsiveness, and infantile seizures permanently damage this plasticity.
Although the median baseline monthly seizure rate was high (48 seizures per month), there was a wide range (2 to 1,650 seizures per month). Thalamo-cortical inhibition and excitation may play important roles in regulation of generalized seizures. Patients with frequent generalized seizures may have widespread deficiencies of thalamo-cortical inhibition, and thus may be amenable to remediation by VNS therapy. In a study employing O15-H20-PET to measure cerebral blood flow in partial-onset epilepsy patients treated with VNS, Henry et al.10 found that the greatest seizure rate reductions occurred in patients who had the greatest increases in thalamic blood flow with VNS.
Appendix
Other members of the E04 VNS Study Group were Debra Sala, MS, PT, Hospital for Joint Diseases/Orthopedic Institute, New York, NY; Robert Burgerman, MD, Sacramento Epilepsy Center, Sacramento, CA; Richard Gilmartin, MD, Research Institute of Kansas, Wichita, KS; Waqar Mirza, MD, Midwest Epilepsy Center, St. Louis, MO; Gershon Ney, Long Island Jewish Medical Center, New Hyde Park, NY; Ivan Osorio, MD, University of Kansas Medical Center, Kansas City, KS; and Martin Salinsky, MD, Oregon Health Science University, Portland, OR.
Acknowledgments
Supported by a clinical research grant from Cyberonics, Webster, TX.
- Received August 28, 1998.
- Accepted in final form January 28, 1999.
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
