CGM Device Guide
Eversense 365
The only implantable CGM. One sensor lasts a full year, sits under the skin with nothing visible on the surface, and calibrates primarily once per week. iCGM accuracy thresholds met in the 365-day ENHANCE trial.
Not currently available in the UK. FDA cleared September 2024 and marketed in the United States. This guide is for reference.
Framework status
Data sufficiency: Met – n=110 adults, 40,497 matched YSI-CGM pairs over 365 days (Bailey et al., 2025, ENHANCE).
Device at a glance
| Manufacturer | Senseonics (distributed by Ascensia Diabetes Care) |
| Wear duration | 365 days per implant – one minor outpatient procedure per year |
| ±20/20 agreement | 93.4% across 40-400 mg/dL – accurate enough to dose from without a confirmatory finger prick |
| Outside ±40/40 | 0.4% of readings – the black swan zone. See the accuracy page. |
| Calibration required | Yes – approximately one finger-prick calibration per week after day 14 (four calibrations on day 1, one per day through day 13) |
| Sensor | Fluorescence-based optical sensor (3.5 mm by 18.3 mm), inserted under local anaesthetic into the upper arm by a trained clinician |
| Transmitter | Rechargeable on-body smart transmitter held in place with a silicone adhesive patch over the implant site. Removable without disturbing the sensor. |
| Alerts | On-body vibratory alerts (unique – silent and discreet) plus smartphone app alerts |
| Display | Eversense companion app (iOS and Android). Eversense Now follower app and Apple Watch integration. |
| Indication | Non-adjunctive (FDA iCGM designation – insulin dosing without confirmatory finger prick) |
| Age indication | Adults 18 years and older (FDA K241335) |
| Sensor survival | 90% of sensors functional at day 365 (Kaplan-Meier, Bailey 2025) |
Accuracy
The ENHANCE study (Bailey et al., 2025) enrolled 110 adults with diabetes across four US clinical sites and generated 40,497 matched CGM-YSI pairs from 14 in-clinic accuracy visits spread across a full year of sensor wear. Overall ±20/20 agreement was 93.4% and ±40/40 agreement was 99.6% – meaning 0.4% of readings fell outside the safety band. iCGM special controls were met across all glucose ranges.
Accuracy was stable across the year. In 30-day intervals, ±20/20 agreement ranged from 90.6% (days 1-30) to 96.0% (days 121-150), with every interval above 89%. This level of consistency across a 365-day implant is the central finding of the trial.
GNL reports CGM accuracy using ±20/20 and ±40/40 agreement only – see the accuracy page for why.
| YSI glucose range (mg/dL) | Matched pairs | ±20/20 agreement | ±40/40 agreement |
|---|---|---|---|
| Overall (40-400) | 40,497 | 93.4% | 99.6% |
| Below 54 | 358 | 95.0% | 99.4% |
| 54-69 | 2,446 | 96.7% | 99.6% |
| 70-180 | 23,130 | 91.4% | 99.4% |
| 181-250 | 7,997 | 94.5% | 99.9% |
| Above 250 | 6,566 | 95.9% | 100.0% |
The confirmed alert detection rate at 70 mg/dL was 96.6%, and at 180 mg/dL was 97.9%. These are hypoglycaemia and hyperglycaemia detection rates at the threshold plus predictive settings – strong performance and a practical reason the on-body vibration alert matters.
The black swan – when to trust and when to check
0.4% of Eversense 365 readings fell outside the ±40/40 accuracy band in ENHANCE – the range where an error is large enough to affect a dosing decision meaningfully. This applies to all framework-qualified CGMs, not only Eversense. See the accuracy page for the precautionary principle and when to reach for a finger prick.
AID system compatibility
iCGM special controls were met in the ENHANCE trial, which is the regulatory gateway for AID integration. However, there is no commercially available AID system that currently uses Eversense 365 as its sensor. Eversense 365 is used as a standalone CGM and cannot be paired with 780G, Control-IQ, Omnipod 5, CamAPS FX, or any other AID system at present.
If AID integration is a priority, Eversense 365 is not the right choice today. The Dexcom G7 offers the broadest current AID compatibility.
Features
Only implantable CGM
The sensor sits in the subcutaneous tissue of the upper arm. Nothing is visible on the skin except the removable transmitter patch – the most discreet CGM form factor available. Insertion and removal are minor office-based procedures under local anaesthetic.
365 days per sensor
One implant per year replaces 24 or more sensor changes with any transcutaneous CGM. Fewer consumables, no adhesive build-up, no scheduled sensor swaps. Of the 110 primary sensors in ENHANCE, 90% survived the full 365 days.
On-body vibratory alerts
The transmitter vibrates on the skin for hypoglycaemia and hyperglycaemia alerts – silent, discreet, and independent of phone volume. Unique to Eversense. Valuable in classrooms, meetings, and shift work where audible alarms are disruptive.
Removable transmitter
The transmitter can be lifted off the skin for charging, swimming, or showering without disturbing the implanted sensor. Daily charging is required. When the transmitter is off, the sensor pauses data transmission but the implant is unaffected.
Weekly calibration
After an initial calibration sequence (four finger pricks on day one, then one per day through day 13), calibration drops to primarily once per week. This is higher than factory-calibrated devices but far less than earlier implantable generations.
Eversense Now and Apple Watch
Follower app allows family members and carers to view real-time glucose data. Apple Watch integration available for on-wrist readings.
Limitations and trade-offs
- Adults only – FDA K241335 is for 18 and older. Eversense 365 is not indicated for children or adolescents.
- No AID integration available – iCGM thresholds met, but no commercial AID system uses Eversense as its sensor.
- Surgical insertion and removal – minor outpatient procedure under local anaesthetic. Not zero-risk. In ENHANCE, 15.5% of participants had a related adverse event, mostly mild and transient (skin infection, wound healing, skin irritation, pain, vasovagal episode).
- Not currently available in the UK – Eversense 365 is FDA cleared (K241335, September 2024) and marketed in the United States. There is no UK access route at time of writing. An insertion-trained clinician is required wherever the device is available.
- Weekly finger-prick calibration – higher calibration burden than factory-calibrated CGMs such as G7, Libre 3 Plus, or Roche SmartGuide.
- Transmitter must be charged daily.
- Year-long commitment – once the sensor is inserted, the choice is fixed until the next replacement. Not the right fit for patients likely to need frequent device changes.
Clinical evidence
- Bailey TS et al. (2025). Evaluation of Accuracy and Safety of the 365-Day Implantable Eversense Continuous Glucose Monitoring System: The ENHANCE Study. Diabetes Technology and Therapeutics 27(5):407-411. n=110 adults, 40,497 matched pairs, ±20/20 93.4% overall, iCGM special controls met. DOI: 10.1089/dia.2024.0592
- FDA K241335 (September 2024). Eversense 365 CGM System clearance. Adults 18 and older. iCGM designation.
- Garg SK et al. (2022). PROMISE Study – Eversense E3 (180-day predecessor). Diabetes Technology and Therapeutics 24(2):84-92.
- Pemberton J et al. (2026). International clinical opinion on transparency, standardisation, and calibration alignment in CGM. Table 1 includes Eversense 365 in the framework. DOI: 10.1111/dom.70460
Who Eversense 365 is right for
Eversense 365 is strongest for adults who want the most discreet CGM form factor, the lowest number of sensor changes, and who do not need AID integration. It is particularly valued by people with adhesive sensitivities or active lifestyles where an on-skin sensor repeatedly gets knocked, and by users who want on-body vibration alerts.
It is not the right choice for children, families seeking AID, or anyone unwilling to accept a minor annual procedure or weekly finger-prick calibration.
Not sure if Eversense 365 is for you? Try the CGM selector or ask Grace.
Evidence backbone
This page is built on the ENHANCE pivotal trial and the GNL CGM framework. John Pemberton is lead or contributing author on the CGM methodology papers below.
- Bailey et al 2025 – ENHANCE Study. Eversense 365 pivotal trial. Diabetes Technology and Therapeutics.
- Pemberton et al 2026 – International clinical opinion on transparency, standardisation, and calibration alignment in CGM. Diabetes, Obesity and Metabolism. DOI: 10.1111/dom.70460
- Pemberton et al 2023 – CGM accuracy: contrasting CE marking with the FDA and TGA. Diabetes, Obesity and Metabolism. DOI: 10.1111/dom.14962
- DSNFUK CGM Comparison Framework v2 (March 2026) – The practical UK clinical reference, authored by John Pemberton. DSNFUK chart
Overall evidence grade: A (pivotal RCT-quality trial with YSI reference, 40,497 matched pairs, iCGM thresholds met) + C methodology layer. Confidence: HIGH.