FDA-approved responsive neurostimulator that records ECoG through four-contact cortical strip and depth leads and delivers stimulation when programmed detections fire. Values come from the FDA summary and the published trials, scoped to the 2013 RNS-300M approval.
Independent, source-linked catalog sheet. Not a manufacturer-issued datasheet, regulatory decision or instructions for clinical use. Human evidence does not establish approval. Source-specific restrictions, conflicts and missing specifications are retained below.
NeuroPace RNS System
The RNS System is an FDA-approved, cranially implanted neurostimulator from NeuroPace that listens to electrocorticography through up to two four-electrode leads and responds with stimulation when a programmed detection fires. This sheet is scoped to the RNS-300M and leads described in the 2013 FDA summary and to the published trials. Later models and supplements are not reviewed.
Identity
Field
Value and source scope
Device
NeuroPace RNS System: RNS-300M neurostimulator (2013 approval) with NeuroPace cortical strip and depth leads; later RNS-320 model named in the 2020 paper [1][2]
Manufacturer
NeuroPace, Inc., Mountain View, California [1]
Interface class
Cranially implanted responsive neurostimulator with one or two cortical strip or depth leads [1]
Feasibility study began 2004 per the long-term paper [2]
First human implant
Unreported
Species studied
Unreported
Regulatory status
FDA PMA P100026, approval November 14, 2013 (panel recommendation February 22, 2013) [1]. Later supplements and the RNS-320 model were not read for this sheet
Function
Records electrocorticography, detects programmed abnormal activity and delivers responsive stimulation; adjunctive therapy for adults with partial onset seizures from no more than 2 foci, refractory to two or more antiepileptic drugs [1]
Target tissue
Seizure foci in or near the brain, via cortical surface strips or stereotactic depth leads [1]
Geometry and architecture
Field
Value and source scope
Interface type
Cortical strip leads on the brain surface and depth leads placed stereotactically; the neurostimulator sits in the cranium coplanar with the skull surface in a ferrule [1]
Array layout
1 x 4 electrode array on each lead; one or two leads per neurostimulator [1]
Electrode count
4 electrodes per lead, up to 2 leads [1]
Pitch
10 mm spacing (cortical strip); 3.5 mm or 10 mm (depth) [1]
Electrode lengths
Lead length 15, 25 or 35 cm (cortical strip); 30 or 44 cm (depth) [1]
Shank width and thickness
Lead diameter 1.27 mm for both lead types [1]
Tip and exposed site geometry
Electrode surface area 0.079 cm2 [1]
Contact coating
Unreported
Insulation
Silicone lead body [1]
Insertion method
Depth leads stereotactic, using a stop gauge to set depth; strip leads placed near epileptic foci; neurostimulator in a craniectomy with ferrule [1]
Anchoring and fixation
Ferrule secures the neurostimulator in the skull; suture sleeves protect the lead body when sutured [1]
Electrode and channel physics
Field
Value and source scope
Exposed site area
0.079 cm2 per electrode [1]
Electrode material
Platinum/iridium [1]
Impedance (with measurement frequency)
Lead conductor resistance listed as 15, 25, 35 ohm (cortical strip) and 30, 44 ohm (depth), +/-10% by lead length; this is lead resistance, not electrode-tissue impedance [1]
Noise floor or SNR
Unreported
Recording modality
Electrocorticographic (ECoG) activity monitored; three programmable detection tools: area, line-length and bandpass [1]
Sampling rate
Unreported
Stimulation capability
Max current 11.5 mA +/-10% and 6 V +/-10% at 500 ohm; pulse width 40-1000 us; 1-333 Hz; 1-1666 pulses per burst; bipolar or multipolar current paths [1]
Charge injection limit
Maximum charge density 25 uC/cm2/phase [1]
Reference and ground
Unreported
Tissue interface and bioresponse
Field
Value and source scope
Target tissue
Seizure foci in cortex or depth structures [1]
Insertion trauma and BBB disruption
Rabbit study (6 days, 4 and 26 weeks): no evidence of systemic toxicity, neurotoxicity or local tissue reaction beyond expected effects of surgical placement and the physical presence of implants [1]
Vascular disruption risk
Unreported
Micromotion sensitivity
Unreported
Gliosis and encapsulation
Rabbit histopathology included GFAP (astroglial activation) and macrophage staining; no reaction beyond expected effects of placement and presence [1]
Neuron loss near sites
Unreported
Foreign-body response mitigation
Unreported
Typical failure modes
RNS-300M: median time to replacement about 1,284 days (3.5 years), with no battery-related device malfunctions; ECoG recording can be affected by radio-frequency identification devices [1][2]
System architecture
Field
Value and source scope
Onboard electronics
Hermetically sealed titanium enclosure with electronic circuitry and a Li-CFx/SVO battery [1]
Data path
Programmer or Remote Monitor wand communicates with the neurostimulator; Remote Monitor uploads data over analog phone lines to the Patient Data Management System [1]
Telemetry bandwidth
Unreported
Sampling rate
Unreported
Power
Li-CFx/SVO primary battery; manual end-of-service estimate 2.6 to 4.2 years depending on settings; RNS-320 anticipated to reach 8 years at moderate use [2]
Thermal management
Unreported
Packaging and hermeticity
Hermetic titanium case; helium leak rate no greater than 5.0 x 10^-9 cc-atm/s per acceptance criteria [1]
MRI compatibility
Contraindicated at the 2013 approval; MR imaging not permitted with any implanted RNS System [1]. Later labeling not read
Surgical complexity
Craniectomy with ferrule, stereotactic or subdural lead placement; infection risk 4.1% per procedure including replacements [2]
Output connectors
Connector cover secures proximal lead contacts to the neurostimulator [1]
Performance envelope
Field
Value and source scope
Acute yield
Unreported
Chronic yield
Unreported
Stability over time
Unreported
Longevity
RNS-300M median replacement about 3.5 years in the long-term study [2]
Revision and explant experience
Serious infection at the implant site in 12.1% of participants; 16 of 35 infections led to explantation [2]
Adverse events
Non-seizure-related hemorrhage in 7 of 256 (2.7%); status epilepticus 8.2%; suicidality-related events 9.8%; 16 deaths over 9 years including probable or definite SUDEP 3.2 per 1,000 patient-implantation years [2]
Notable demonstrations
Class I pivotal trial and 9-year follow-up with 1,895 patient-implantation years [2][3]
Clinical and preclinical evidence
Field
Value and source scope
Human subjects
Feasibility n=65; pivotal n=191; long-term study enrolled 230 of 256 treated [1][2]
Preclinical cohort
Rabbit neuroimplantation and chronic toxicity study [1]
Follow-up duration
Pivotal 2 years; long-term study to 9 years (median follow-up 8.97 years) [2]
Indications
Adults 18 and older, partial onset seizures, no more than 2 foci, refractory to two or more antiepileptic drugs, averaging 3 or more disabling seizures per month [1]
Trials and registries
IDE feasibility G010288; pivotal and long-term treatment studies per the SSED [1]
Primary outcomes
Pivotal blinded period: seizures reduced 37.9% (n=97) with stimulation versus 17.3% (n=94) with sham, p=0.012. At 9 years: median reduction 75%, responder rate 73% [2][3]
Key limitations
Long-term study was open label (Class IV evidence); SSED sections read here stop before its efficacy results; later device models and labeling not reviewed [1][2]
Engineering tradeoffs
Field
Value and source scope
Strengths
Closed-loop detection and stimulation from cortical or depth electrodes with stored ECoG [1]
Limitations
Short battery life of the RNS-300M, infection risk per procedure, MRI contraindication at approval [1][2]
Scaling constraints
Up to 8 electrodes (2 leads x 4) per neurostimulator [1]
Version boundary
The FDA summary describes the RNS-300M neurostimulator and its lead specifications. The 9-year paper names a newer RNS-320 model but gives only an anticipated battery figure. Results listed here belong to the cohorts in the cited papers, not to every later device.