Closed-loop brain stimulation for epilepsy

In plain language

Epilepsy is a condition in which groups of brain cells sometimes fire in sudden, excessive bursts that cause seizures. For some people, medicines do not stop them. One approach is an implanted device that watches the brain's electrical activity and sends a small pulse of stimulation when it sees a pattern that a clinician has programmed in, instead of stimulating on a fixed timetable.

The NeuroPace RNS System is one example. The US Food and Drug Administration (FDA) approved it in November 2013 for adults whose partial-onset seizures, which start in one or two areas of the brain, average three or more disabling seizures a month. Studies show that seizures fall on average, but the results vary widely between people. The device is not a cure, and many people still have seizures.

Technical explanation

The RNS System places electrodes on the cortex over one or two seizure foci and records local electrical activity. The clinician programs an onboard detector to recognise specific patterns. When one is found, the implanted neurostimulator delivers brief stimulus pulses. Because stimulation follows detected activity rather than a timer, the system is described as closed-loop. In the long-term study reported by Bergey and colleagues, the median preimplant baseline was 10.2 disabling seizures per month.

The main evidence comes from the RNS pivotal trial. In that double-blinded, sham-controlled study, 191 implanted adults were randomised 1:1 to active or sham stimulation. Seizure frequency fell by 37.9% with active stimulation and by 17.3% with sham over the blinded period (p = 0.012). In the open-label phase that followed, the median reduction was 44% at one year and 53% at two years, but without a control group. Bergey and colleagues followed 230 participants into a long-term study, with a mean follow-up of 5.4 years. Median reductions were 60% at the start of year 3 and 66% at the start of year 6, while responder rates (50% or more reduction) were 58% and 59% at those points. The most common serious device-related event was implant-site soft-tissue infection (9.0%), and neurostimulator explantation occurred in 4.7%.

For comparison, the SANTE trial of deep brain stimulation of the anterior thalamus reported a median fall of 40.4% in stimulated patients and 14.5% in controls over its blinded phase, and a median reduction of 56% at two years. The sources reviewed for this profile did not include peer-reviewed seizure-reduction results for adaptive deep brain stimulation in epilepsy, so RNS figures should not be assumed to transfer to it.

What the evidence says

Each statement is typed by the strength of its evidence and linked to the source that supports it.

Limitations and open problems

• The blinded comparison in the pivotal trial lasted 12 weeks, and the two-year figures come from an open-label phase with no control group, so regression to the mean and expectation effects cannot be ruled out (Heck and colleagues, 2014; Bergey and colleagues, 2015).
• The main studies were sponsored by NeuroPace. Several pivotal-trial authors report NeuroPace employment or equity, and the long-term study was funded and designed by NeuroPace (Heck and colleagues, 2014; Bergey and colleagues, 2015).
• The approved population is narrow: adults with one or two foci and an average of at least three disabling seizures per month. The device has not been evaluated in people with less frequent seizures, and the pivotal trial enrolled people who were not considered current epilepsy surgery candidates (AES release, 2013; Heck and colleagues, 2014).
• The long-term study reported a sudden unexpected death in epilepsy (SUDEP) rate of 3.5 per 1,000 patient-implant-years (Bergey and colleagues, 2015).
• In the two-year pivotal period, four deaths were attributed to possible or definite SUDEP, and six deaths occurred in total (Heck and colleagues, 2014).
• This profile found no peer-reviewed seizure-reduction figures for adaptive deep brain stimulation in epilepsy.

Possible future uses

Timeline

1. May 2010Fisher and colleagues publish the SANTE trial of deep brain stimulation of the anterior thalamus for refractory epilepsy in Epilepsia, a comparator for closed-loop approaches. Sources: [3]
2. Nov 2013The FDA grants premarket approval to the NeuroPace RNS System for adults with partial-onset seizures that average at least three disabling seizures a month. Sources: [4]
3. Mar 2014Heck and colleagues publish the final two-year results of the RNS pivotal trial in Epilepsia, reporting a 37.9% vs 17.3% blinded-period reduction. Sources: [1]
4. Feb 2015Bergey and colleagues publish long-term RNS outcomes in Neurology, with a mean follow-up of 5.4 years across 230 participants in the long-term study. Sources: [2]

Futures this could help bring about

How far away?

Brain implants that stop seizures before they spread

Scaling up · 5 of 8 steps done

Glossary

Closed-loop stimulation

A device that monitors brain activity and delivers stimulation in response to a detected pattern, rather than on a fixed schedule.

Sham stimulation

A control condition in which the device is implanted but does not deliver active therapy, used to separate real treatment effects from expectation effects.

Premarket approval

The FDA's most stringent route for approving a high-risk medical device before it can be sold in the United States.

SUDEP

Sudden unexpected death in epilepsy, a rare death in people with epilepsy where no other cause is found.

Sources

1. peer reviewed paper
   Two-year seizure reduction in adults with medically intractable partial onset epilepsy treated with responsive neurostimulation: Final results of the RNS System Pivotal trial
   Christianne N. Heck, David King-Stephens, Martha J. Morrell. Epilepsia, Mar 2014. DOI: 10.1111/epi.12534. Accessed 21 Oct 2026.
2. peer reviewed paper
   Long-term treatment with responsive brain stimulation in adults with refractory partial seizures
   G. K. Bergey, M. J. Morrell, E. M. Mizrahi. Neurology, 24 Feb 2015. DOI: 10.1212/WNL.0000000000001280. Accessed 21 Oct 2026.
3. peer reviewed paper
   Electrical stimulation of the anterior nucleus of thalamus for treatment of refractory epilepsy
   Robert Fisher, Vicenta Salanova, Thomas Witt et al.. Epilepsia, May 2010. DOI: 10.1111/j.1528-1167.2010.02536.x. Accessed 21 Oct 2026.
4. news
   Brain stimulation device demonstrates safety, seizure reduction
   ScienceDaily (American Epilepsy Society release), 10 Dec 2013. Accessed 21 Oct 2026.

Revision history

3 changes since this profile was created.