
Libre 2 produced acceptable clinical error grids in ICU testing, but neither glucose method met ISO accuracy criteria
AI-summarized from the linked source. Educational brief, not medical advice.
Brief summary
In 157 paired ICU measurements, FreeStyle Libre 2 had a mean absolute relative difference similar to capillary testing and placed all readings in clinically acceptable error-grid zones, but neither method met ISO numerical accuracy criteria and hypoglycemic measurements were limited.
What NurseJet pulled from the source
This diagnostic accuracy study compared FreeStyle Libre 2 interstitial glucose and StatStrip capillary glucose with central laboratory venous or arterial glucose in critically ill adults receiving insulin plus ventilatory or vasopressor support. Across 157 paired measurements, mean absolute relative difference was 10.43% for Libre 2 and 7.58% for capillary glucose, a nonsignificant difference. Neither method met ISO 15197:2013 numerical accuracy criteria. Libre 2 placed 100% of readings in zones A and B on both Clarke and Parkes clinical error grids; capillary glucose placed 97.6% and 100%, respectively. Few hypoglycemic events limit conclusions in the range where confirmation is especially safety critical.
Why this matters for nurses
Critical-care nurses titrate insulin and respond to glucose trends in patients whose perfusion, edema, medications, and rapid physiologic changes may affect bedside measurements. The study supports cautious trend use while reinforcing laboratory or protocol-directed confirmation when readings are discordant or clinically consequential.
Bedside takeaway
Libre 2 trends may be clinically useful in selected ICU patients, but low or discordant values still require protocol-directed confirmation.
How This Applies in Practice
Use this when: Monitoring an ICU patient receiving insulin with a facility-approved continuous glucose system.
On your shift
- Assess the glucose trend alongside perfusion, medications, nutrition, and the patient's current clinical condition.
- Inspect the sensor and obtain the required confirmatory sample when a value is low, rapidly changing, unexpected, or inconsistent with symptoms.
- Document the device reading, confirmatory result, and response when escalating a clinically important discrepancy.
Explain this for my unit
Key takeaways
- The study evaluated 157 paired measurements against central laboratory glucose in critically ill adults receiving insulin.
- Libre 2 and capillary glucose had comparable mean absolute relative difference in this sample.
- Neither monitoring method met ISO 15197:2013 numerical accuracy criteria.
- All Libre 2 readings fell in clinically acceptable error-grid zones, but too few hypoglycemic events were captured to establish low-range reliability.
Practice implications
- Use continuous glucose data only within the ICU's approved insulin and confirmation protocol. Compare unexpected readings with the patient's condition and trend, inspect the sensor site, and obtain the required confirmatory sample before a treatment change when values are low, rapidly changing, or discordant.
Limitations & cautions
- The abstract reports 157 paired measurements but not the number of participants or all clinical subgroups. Few hypoglycemic measurements were available, both index methods missed the ISO numerical standard, and error-grid placement does not establish interchangeability with laboratory testing. The findings may not generalize to other sensors, perfusion states, or insulin protocols.
- AI-summarized from the linked source. Review the original article before applying to practice.
Citations
Exact source links
Public citations are filtered to exact credible source pages. Homepage-only or invalid links stay in admin review and are not shown here.
Clinical medicine insights. Endocrinology and diabetes (PubMed)
Clinical medicine insights. Endocrinology and diabetes (PubMed). Numerical and Clinical Accuracy of the FreeStyle Libre 2 System: A Diagnostic Accuracy Study in Critically Ill Patients.
https://pubmed.ncbi.nlm.nih.gov/42473549/
Professional education only


