Long-term efficacy and safety of thalamic stimulation for drug-resistant partial epilepsy

Abstract

Objective: To report long-term efficacy and safety results of the SANTE trial investigating deep brain stimulation of the anterior nucleus of the thalamus (ANT) for treatment of localization-related epilepsy.

Methods: This long-term follow-up is a continuation of a previously reported trial of 5- vs 0-V ANT stimulation. Long-term follow-up began 13 months after device implantation with stimulation parameters adjusted at the investigators' discretion. Seizure frequency was determined using daily seizure diaries.

Results: The median percent seizure reduction from baseline at 1 year was 41%, and 69% at 5 years. The responder rate (≥50% reduction in seizure frequency) at 1 year was 43%, and 68% at 5 years. In the 5 years of follow-up, 16% of subjects were seizure-free for at least 6 months. There were no reported unanticipated adverse device effects or symptomatic intracranial hemorrhages. The Liverpool Seizure Severity Scale and 31-item Quality of Life in Epilepsy measure showed statistically significant improvement over baseline by 1 year and at 5 years (p < 0.001).

Conclusion: Long-term follow-up of ANT deep brain stimulation showed sustained efficacy and safety in a treatment-resistant population.

Classification of evidence: This long-term follow-up provides Class IV evidence that for patients with drug-resistant partial epilepsy, anterior thalamic stimulation is associated with a 69% reduction in seizure frequency and a 34% serious device-related adverse event rate at 5 years.

Figure 1. Participant timeline

The number of subjects who completed, or discontinued before, each visit is indicated in the figure. Reasons for discontinuation between phases are as follows. ^aPreimplant discontinuations: have been described in detail by Fisher et al. ^bOne year: device explant (4: implant site infection in 2 subjects, discomfort, involuntary muscle contractions); SUDEP (1). ^cTwo years: device explant (2: implant site infection, therapeutic product ineffective); drowning (1). ^dThree years: device explant (3: anxiety, cognitive disorder, meningitis); withdrawal of consent (1). ^eFour years: device explant (4: therapeutic product ineffective in 2 subjects, psychotic disorder, undesirable change in stimulation); completed suicide (1); physician choice (1). ^fFive years: device explant (5: therapeutic product ineffective in 4 subjects, implant site infection); withdrawal of consent (2); SUDEP (1); physician choice (1). ^gMore than 5 years: device explant (4: anxiety, convulsion, implant site infection, therapeutic product ineffective); withdrawal of consent (2); cardiac arrest (1); liver cancer (1). ^hSubjects have been followed through 6 (80 subjects), 7 (41), 8 (31), and 9 (6) years. *One hundred nine of 110 implanted subjects were randomized, but all subjects continued to be followed. Statistical imputation is based on 109 randomized subjects.

Figure 2. Percentage seizure reduction over time

The graph shows seizure reduction for those subjects who had at least 70 days of diary in the 3 months before each annual visit (blue bars) as well as sensitivity analyses allowing diaries as short as 28 days, and using either last observation carried forward (red bars) or a worst case (100% worsening from baseline, yellow bars) data imputation methodology for subjects with fewer than 28 diary days. Also shown is seizure reduction for those subjects who had at least 28 days of diary in the 3 months before each and every annual visit, elucidating whether, in the most diary-compliant subgroup of subjects, seizure reductions change over time (constant cohort, green bars). *Wilcoxon signed-rank, p < 0.001; **25th percentile bar extends to 100% since more than 25% of 109 subjects were imputed to or had greater than 100% median percent change in total seizure frequency from baseline.

Figure 3. Distribution of individual subject response to treatment

The graph shows the by-subject distribution of total seizure frequency percent change from baseline at 5 years for subjects who had at least 70 days of diary in the 3 months before the year 5 visit. Negative values indicate a seizure frequency reduction compared with baseline.

Figure 4. Seizure severity and quality of life

The graph shows responses to the Liverpool Seizure Severity Scale and the 31-item Quality of Life in Epilepsy (QOLIE-31) scale for all subjects who completed the respective questionnaire at each annual visit. Higher values reflect improvement in both charts. A QOLIE-31 change of at least 5 points is considered clinically meaningful. *Paired t test, p < 0.001.

Figure 5. Neuropsychological outcomes

The graph shows composited T scores across several neuropsychological tests within each given domain, over time. Positive values represent an improvement in each outcome. *p < 0.001, **p < 0.05, Wilcoxon signed-rank nonparametric test.