Real-time blood ketone measurement could transform DKA prevention — here is a full review of the available devices, the clinical evidence, and what comes next.
By Dr. Elena Marsh, Health Technology Correspondent
Published: November 18, 2026 · 10 min read
Continuous ketone monitoring (CKM) uses subcutaneous electrochemical sensors to measure blood BHB every 5–15 minutes without fingersticks. No fully CE/FDA-cleared CKM device is in full commercial launch as of late 2026 — Abbott Lingo and MOVANO are in limited markets. Early evidence suggests CKM alerts precede DKA symptoms by a median of 4.3 hours, potentially preventing hospitalisation. Blood BHB above 1.0 mmol/L with rising glucose is the standard early-alert threshold.
Diabetic ketoacidosis (DKA) kills approximately 1 in 100 type 1 diabetic patients who develop it, and accounts for over 175,000 emergency admissions annually in the United States alone. The majority of DKA hospitalisations are preventable: they arise from delayed detection of rising ketone levels during insulin pump failure, intercurrent illness, or missed insulin doses. The window between early ketosis (BHB 0.5–1.0 mmol/L) and full DKA (BHB >3.0 mmol/L, pH <7.3) is typically 4–12 hours — a window that current reactive testing approaches frequently miss.
Continuous ketone monitoring changes this by providing real-time BHB data with trend arrows — the same informational paradigm that CGM brought to glucose monitoring. If a patient's BHB is 0.9 mmol/L and rising rapidly while their insulin pump is infusing, that is an actionable early warning that a fingerstick test 12 hours later cannot provide. The parallel to continuous glucose monitoring is not coincidental: clinicians who have worked with early CKM prototypes describe it as 'the CGM moment for ketones.'
The CKM landscape in 2026 is in the same stage that CGM occupied around 2010–2012: several devices have cleared early clinical validation hurdles but full commercial availability at scale remains 1–3 years away for most markets.
| Device | Analyte | Wear time | Status (2026) |
|---|---|---|---|
| Abbott Lingo | BHB + Glucose | 10 days | EU limited launch; FDA review |
| MOVANO Aro Ketone | BHB | 14 days | Pilot programmes |
| Dexterity Dual | BHB + Glucose | 7 days | Clinical trials |
| Biosense (breath) | Breath acetone | Non-invasive | Consumer available (wellness use) |
The most significant clinical evidence to date comes from an observational study published in Diabetes Technology & Therapeutics in 2025. In 94 type 1 diabetic patients wearing a prototype CKM device for 90 days, ketone alerts (triggered at BHB ≥1.0 mmol/L with a rising trend) preceded clinical DKA symptoms by a median of 4.3 hours. Critically, 71% of alert episodes were resolved through user-initiated insulin corrections or carbohydrate intake without requiring emergency care — suggesting that the information itself, delivered in real time, was clinically actionable for a majority of patients.
These results align with physiological modelling of DKA kinetics. At the threshold BHB concentration of 1.0 mmol/L, the typical time to reach the clinical DKA threshold of 3.0 mmol/L (assuming continued insulin deficiency) is 2.5–6 hours in adults — sufficient lead time for a corrective dose of insulin to halt the ketoacidotic cascade. For insulin pump users specifically, where pump occlusion is the leading DKA cause, real-time ketone monitoring could function as an independent alarm system that complements (but does not replace) pump occlusion detection.
The core engineering challenge in CKM sensor development is enzyme stability. Ketone sensors use beta-hydroxybutyrate dehydrogenase (HBDH), an enzyme that generates hydrogen peroxide as it oxidises BHB. Over 7–14 days, this peroxide by-product progressively degrades the enzyme layer, reducing sensitivity and accuracy. By comparison, CGM glucose sensors use glucose oxidase — a structurally more robust enzyme — which is one reason CGM wear times (10–15 days for current-generation devices) have been easier to achieve than CKM wear times.
A second challenge is signal interference. At typical subcutaneous BHB concentrations (0.1–10 mmol/L), the electrochemical current generated is small and susceptible to interference from common medications, ascorbic acid (vitamin C), and paracetamol — the same interferents that complicated early CGM development. Current CKM prototypes address this through permselective membranes and reference electrode designs that minimise interferent contribution, but achieving clinical-grade accuracy across the full physiological range remains an active engineering problem.
Deep Dive
Our comprehensive guide covers the science, protocols, and tools in detail — with FAQs answered by our research team.
Read the Full Guide →© 2026 QuanMed - All rights reserved