Devices

A fully implanted endovascular BCI: a stent-electrode array in the superior sagittal sinus recording ECoG-like signals, trading spikes for catheter-based deployment.

Device — Endovascular

Stentrode (Synchron)

Official site →

BCI · endovascular · Stentrode · Synchron · ECoG-like · minimally invasive

Stentrode (Synchron)

One-line verdict: A fully implanted endovascular BCI that records ECoG‑like signals from inside the superior sagittal sinus using a stent‑electrode array—trading intracortical spike fidelity for a catheter-based, no‑craniotomy implantation path.

Quick tags: Recording · Stimulation: unknown/limited · Channels: 16 · Species: Human + sheep · First implanted: 2019–2020 era

Overview

What it is: A self‑expanding nitinol stent with integrated electrodes deployed transvenously into the superior sagittal sinus adjacent to motor cortex, connected by a lead to a subcutaneous implant that relays data to an external unit.

Why it matters: It’s the flagship example that you can get chronic, useful motor intent signals without penetrating cortex and without a craniotomy.

Most comparable devices: Subdural ECoG grids (signal class), intracortical arrays (intent use‑case), other endovascular BCIs (design family).

Spec Card Grid

Identity

  • Device name: Stentrode (stent‑electrode array)
  • Canonical ID: BTSD-0003
  • Inventor / key authors: Thomas Oxley et al. (clinical translation)
  • Org / manufacturer: Synchron
  • First demonstrated (year): preclinical large‑animal work preceded first‑in‑human (dates vary by paper)
  • First implanted (year): first‑in‑human experience published online Oct 28, 2020 (implantation occurred prior)
  • Species: Human; large‑animal (ovine) model used heavily
  • Regulatory / trial status: Human clinical studies (e.g., first‑in‑human; SWITCH)
  • Primary use: Recording (motor intent → digital control)
  • Primary target: Motor cortex adjacency via superior sagittal sinus

Geometry & Architecture

  • Interface type: Endovascular cortical interface (venous)
  • Penetrating?: no (does not penetrate cortex; sits in a blood vessel)
  • Form factor: self‑expanding nitinol stent with mounted electrodes
  • Array layout: linear distribution along the stent body (functional “strip” along the sinus)
  • Scaffold size: 8 × 40 mm (reported)
  • Electrode count: 16
  • Site spacing (µm): ~3000 (reported as ~3 mm)
  • Insertion method: transvenous catheter delivery via internal jugular vein → superior sagittal sinus deployment
  • Anchoring method: stent apposition + endothelialization over time
  • Packaging location: lead tunneled subcutaneously to a chest (subclavicular) telemetry implant pocket

Electrode & Channel Physics

  • Channel count: 16
  • Active sites used (vs total): typically all 16 are available; actual use can depend on signal quality
  • Electrode material: platinum electrodes on nitinol scaffold (reported)
  • Site area (µm²): not consistently reported in open summaries
  • Electrode size: 500 µm diameter (reported)
  • Impedance @ 1 kHz / noise floor: not consistently reported in open clinical summaries (mark unknown)
  • Recording modality: vascular ECoG‑like signals; spectral motor intent features reported
  • Stimulation capability: unknown for this catalog unless tied to a primary device‑specific stimulation report
  • Charge injection limit / safe stim range: not publicly disclosed

Tissue Interface & Bioresponse

  • Target tissue: cortex adjacency via venous wall
  • BBB disruption: likely lower than penetrating arrays (non‑penetrating placement), but treat as low–moderate unless quantified
  • Vascular disruption risk: thrombosis/stenosis/migration are the dominant device-class risks and are monitored in studies
  • Micromotion sensitivity: different from shanks; vessel pulsatility introduces motion but tissue penetration is avoided
  • Gliosis / encapsulation: not the same failure mode as penetrating shanks; coupling can change with vessel wall remodeling
  • Typical failure mode: venous remodeling affecting coupling, system-level lead/telemetry issues, decoder drift

System Architecture

  • Onboard electronics: implanted telemetry unit connected to the stent via a lead
  • Data path: implanted internal hardware + external telemetry + computer interface chain (no percutaneous skull pedestal)
  • Telemetry method: reported as infrared optical transmission between external/internal telemetry units in early descriptions
  • Power: external unit inductively powers the implanted telemetry unit (reported)
  • Hermeticity: implantable package (details vary by generation)
  • MRI compatibility: unknown/conditional pending manufacturer documentation
  • Surgical complexity: neurointerventional catheter procedure (angiography suite), no craniotomy

Performance Envelope

  • What it enables: click/switch-like command control + OS interaction using motor intent; assistive stacks often include eye tracking
  • Signal class: best for robust spectral features (including gamma/high‑gamma), not single‑unit spikes
  • Longevity: follow‑up reported to 12 months in early cohorts; broader longevity still emerging

Clinical / Preclinical Evidence

  • Early feasibility (human): first‑in‑human home use described in severe paralysis cohorts
  • N implanted subjects / animals: small cohorts in early human reports; ovine model used preclinically
  • Trial registry links: to add (SWITCH / related studies)
  • Primary outcomes: feasibility of computer control for activities of daily living tasks
  • Key limitations of evidence: small N, heterogeneous assistive stacks, limited open hardware-spec reporting

Engineering Verdict

Strengths:

  • catheter-based deployment (no craniotomy)
  • non‑penetrating cortical tissue interface class
  • avoids percutaneous skull connectors

Limitations / failure modes:

  • lower spatial resolution than intracortical spikes (16 macroelectrodes)
  • signal coupling depends on vascular wall + remodeling
  • channel count scaling constrained by venous anatomy + safe stent geometry

Scaling constraints:

  • venous patency/thrombosis risk envelope
  • electrode density limited by stent mechanics + hemodynamics
  • telemetry/power pipeline must stay within implant heat and regulatory limits

What it’s trying to fix vs Utah / N1:

  • surgical invasiveness
  • percutaneous infection risk
  • micromotion damage from rigid penetrating shanks

Simulation Hooks (for BuildTheSimulation)

  • Minimal model to reproduce: stent electrode inside a compliant venous tube adjacent to cortex; coupling transfer function from cortical sources → vessel wall → electrode
  • Parameters to expose as sliders: stent diameter/oversizing proxy, electrode spacing/count, vessel-wall thickness/neointimal growth (attenuation), source depth/orientation
  • What outputs to visualize: bandwidth/SNR proxy vs anatomy, long‑term attenuation vs neointimal thickness, patency risk proxy vs oversizing

References