Controlled Release Emergency Seizure Therapy — an electronics-free, pH-triggered subcutaneous implant concept for autonomous seizure rescue.
No prototype. No animal data. No clinical claims. This is a pre-prototype concept note filed on Zenodo to establish date of priority and invite technical critique.
About 30% of people with epilepsy have drug-refractory disease. For this group, a generalised tonic-clonic seizure outside hospital — no caregiver, no rescue medication — can progress to status epilepticus in minutes.
Existing closed-loop devices (NeuroPace RNS, VNS) work, but cost $15–40k USD, require cranial surgery, and run on batteries that need periodic replacement. Nothing in that price range is available for most of the world.
The author developed status epilepticus and was admitted to intensive care. This project is a direct response to the absence of an affordable autonomous rescue option.
A small subcutaneous capsule (~25 mm) that uses the lactic acidosis produced during a sustained seizure as a purely mechanical chemical trigger — no electronics, no battery, no specialist programming.
| Layer | Material | Function |
|---|---|---|
| Outer shell | Medical-grade silicone, microporous face | Lets H⁺ ions equilibrate with tissue |
| Sentinel buffer | Bicarbonate solution | Absorbs short-duration acidosis (exercise) |
| Polymer barrier | Eudragit S100 | Erodes when pH < 6.95 — after buffer is exhausted |
| Drug reservoir | Midazolam + atropine | Released through rate-controlled exit membrane |
A naive pH threshold would false-trigger on HIIT or competitive sport. The sentinel buffer solves this by acting as a temporal integrator: brief exercise-grade acidosis (10–15 min) is absorbed without reaching the polymer; sustained seizure-grade acidosis (30–60+ min) exhausts the buffer locally and triggers erosion.
The concept is grounded in published physiology but the central link is unverified:
| Evidence | Source | Directness |
|---|---|---|
| Arterial pH 7.14 ± 0.06 post-seizure, normalises in 60 min | Orringer et al., NEJM 1977 | Direct human data (blood) |
| Brain ECF pH drop ~0.36 units, persists ≥45 min post-ictal | Siesjö et al., JCBFM 1985 | Animal data (brain ECF) |
| Blood-to-interstitial lag ~2–8 min | CGM glucose literature | Indirect analogy |
| Subcutaneous pH during GTCS in humans | Not found in literature | This is the gap |
The primary experimental question before any prototype work: does subcutaneous tissue pH drop below 7.0 during a human GTCS, and for how long?
- Subcutaneous pH kinetics during human GTCS (the critical unknown)
- Eudragit S100 erosion rate in subcutaneous conditions vs. GI conditions
- Midazolam stability in polymer matrix at 37°C over 12–24 months
- Foreign body response effect on pH equilibration and drug release
- Sentinel buffer sizing to set the exercise/seizure discrimination threshold
crest-implant/
├── README.md
├── CITATION.cff
├── LICENSE
├── CREST_Concept_Note_v2.pdf ← full concept paper with references
└── figures/
├── figure1_cross_section.svg ← vector source
├── figure1_cross_section.png
├── figure2_timeline.svg
└── figure2_timeline.png
@misc{yasenovyi2026crest,
author = {Yasenovyi, Varfolomii},
title = {{CREST}: an electronics-free, pH-triggered subcutaneous implant
for emergency seizure rescue},
year = {2026},
publisher = {Zenodo},
doi = {10.5281/zenodo.19976598},
url = {https://doi.org/10.5281/zenodo.19976598}
}Text and figures: CC BY 4.0
Use freely, cite the Zenodo DOI.
Technical critique is the point of making this public. If you know polymer erosion kinetics, seizure physiology, or subcutaneous drug delivery — open an issue or email via GitHub profile.