Danish Review: CGM Accuracy in ICU Patients Varies Widely

A Danish Review Questions CGM Reliability for Critically Ill Patients

A study from Denmark, published in Intensive Care Medicine (2024), takes a sweeping look at a question that matters for anyone with diabetes who may one day end up in a hospital intensive care unit (ICU): Can you trust a continuous glucose monitor (CGM) when you're critically ill?

Researchers at Copenhagen University Hospital–North Zealand reviewed 96 studies published through May 2024, making this one of the most comprehensive international examinations of CGM accuracy in ICU settings to date. While most CGM research focuses on outpatient use — people managing diabetes at home — this Danish team zeroed in on the high-stakes hospital environment where blood sugar control can be a matter of life or death.

What the Researchers Found

The short answer: CGM accuracy in the ICU is highly variable — and sometimes alarmingly poor.

• The key accuracy metric used was MARD (Mean Absolute Relative Difference) — essentially, how far off the CGM reading is from a true blood sample result, expressed as a percentage. Lower is better.
• For the most common type of CGM (worn under the skin, or subcutaneous), MARD ranged from 6.6% to 30.5% across studies. A 30% error on a blood sugar reading of 180 mg/dL, for example, could mean the device shows anywhere from 126 to 234 mg/dL — a clinically dangerous gap.
• Newer factory-calibrated CGMs (the kind most US patients use today, like Dexcom G7 or FreeStyle Libre 3) showed a narrower but still significant MARD range of 9.7% to 20.6% in the ICU.
• Intravascular CGMs — devices placed directly into a vein or artery, not yet widely available for home use — performed better, with MARD as low as 5%.
• Accuracy appeared to be worst during hypoglycemia (low blood sugar), which is precisely when accurate readings matter most in critical care.

Why the ICU Environment Is Different

This international perspective is especially relevant because ICU conditions — poor circulation, IV fluids, medications like vasopressors, and rapid metabolic changes — can all interfere with how CGM sensors read glucose in the fluid just beneath the skin. Most US-based CGM studies are conducted on stable outpatients, so Danish researchers examining the ICU context fills an important gap in the evidence base.

How This Compares to US Guidelines

Current US guidelines from the American Diabetes Association (ADA) and the Society of Critical Care Medicine acknowledge that glycemic control in the ICU is important but do not yet broadly recommend CGM as a replacement for standard blood glucose testing in critically ill patients. This Danish review aligns with that cautious stance and suggests that more standardized research protocols are needed before CGM can be confidently adopted in ICU settings.

Why This Matters for US Patients

If you or a loved one with diabetes is ever hospitalized in an ICU, it's important to know that your CGM — even a top-of-the-line device — may not be as accurate as it is at home. Advocating for frequent confirmatory blood draws is reasonable and supported by this research. For day-to-day diabetes management outside the hospital, CGM remains a proven tool. MDS Diabetes (mdsdiabetes.com) offers a range of CGM-compatible supplies and glucose monitoring accessories to help you stay accurate in everyday life, where these devices perform at their best.

Full Citation

[Denmark] — Nielsen, Grigonyte-Daraskeviciene, Olsen (2024). Accuracy of continuous glucose monitoring systems in intensive care unit patients: a scoping review. Intensive Care Medicine. PMID: 39417874. DOI: 10.1007/s00134-024-07663-6. Institution: Department of Anesthesiology and Intensive Care, Copenhagen University Hospital–North Zealand, Hilleroed, Denmark.