Efficacy of vagus nerve stimulation as a treatment for medically intractable epilepsy in brain tumor patients. A case-controlled study using the VNS therapy Patient Outcome Registry

Abstract

Purpose: Vagus nerve stimulation (VNS) therapy is a procedure to control seizure frequency in patients with medically intractable epilepsy. However, there is no data on efficacy in the subset of these patients with brain tumors. The purpose of this study is to evaluate the efficacy of VNS therapy in patients with brain tumor-associated medically intractable epilepsy.

Methods: Data from the VNS therapy Patient Outcome Registry, maintained by the manufacturer of the device, Cyberonics Inc. (Houston, TX, USA), was queried to characterize the response of patients in whom a brain tumor was listed as the etiology of epilepsy. A case-control analysis was implemented and patient outcome was measured by Engel classification, median seizure response and responder rate (≥50% seizure reduction) using t-tests and chi-squared tests.

Results: In 107 patients with an epilepsy etiology related to a brain tumor, seizure reduction was 45% at 3 months and 79% at 24 months with a responder rate of 48% at 3 months and 79% at 24 months. There was no statistical difference in seizure reduction compared with 326 case-control patients from the registry without brain tumors. There was no significant difference in anti-epileptic drug (AED) usage from baseline to 24 months post implant in either group.

Conclusions: VNS therapy is equally effective in patients who suffer seizures secondary to brain tumors as in patients without history of a brain tumor. VNS therapy is a viable treatment option for patients with brain tumor associated medically intractable epilepsy, assuming cytoreductive and other adjuvant therapies have been fully explored.

Keywords: Brain tumors; Medically intractable epilepsy; Seizures; Vagus nerve stimulation.

Fig. 1

Study design summary. Data pertaining to patients with and without brain tumor history were extracted from 7383 total individuals in the VNS therapy Patient Outcome Registry (A). Individuals without a specified etiology or brain tumor history were excluded from the study, as well as those without any data on follow-up visits. Solid lines indicate included patients; dashed lines indicate excluded patients. Of the 114 brain tumor patients with a known epilepsy etiology related to the tumor, 107 patients had at least one follow-up visit matched to two non-brain tumor patient follow-up visits (B).

Fig. 2

Raw data depicting seizure outcomes after VNS therapy in brain tumor patients (BTP) versus non-BTP. Median percent decrease in seizure frequency (A) and responder rates (B) are shown for BTP and non-BTP patients after VNS therapy at 3, 6, 12, and 24 months. Engel outcomes are shown at 3 months (C) and 24 months (D) after VNS therapy. For (A–D), statistical analysis is deferred given unequal samples (see Table 1). N = 93, 73, 55, and 12 for BTP and 1509, 1004, 943, and 390 for non-BTP at 3, 6, 12, and 24 months, respectively.

Fig. 3

Case–control design data depicting seizure outcomes after VNS therapy in BTP versus non-BTP. Median percent decrease in seizure frequency (A) and responder rates (B) are shown for BTP and non-BTP patients after VNS therapy at 3, 6, 12, and 24 months. Engel outcomes are shown at 3 months (C) and 24 months (D) after VNS therapy. Statistical significance (P < 0.05) was not reached at any time point between BTP and non-BTP for any seizure outcome measure. For A and B, statistical analysis reflects Wilcoxon Rank-Sum test and two-proportion z-test, respectively. For (C and D), statistical analysis reflects Cochran–Mantel–Haenszel statistic for row means scores. N = 86, 70, 52, and 19 for BTP and 172, 140, 104, and 38 for non-BTP at 3, 6, 12, and 24 months, respectively.

Fig. 4

Baseline seizure types. Baseline seizure types are shown for BTP versus non-BTP using raw data (A) and following case–control selections (B). Many patients presented with multiple seizure types, so each seizure type was analyzed separately using chi-square tests to show that no difference existed between the groups after matching. *Statistically significant value (P < 0.05). CPS, complex partial seizure; GTC, generalized tonic-clonic seizure; second-GTC, secondarily generalized tonic-clonic seizure; SPS, simple partial seizure.