Dexcom G7 Accuracy and Reproducibility in the Intensive Care Unit

Area of Science:

• Medical Devices
• Endocrinology
• Critical Care Medicine

Background:

• Continuous glucose monitoring (CGM) devices are increasingly used for diabetes management.
• Accuracy of CGM devices in intensive care unit (ICU) settings requires specific evaluation due to unique patient physiology and clinical practices.

Purpose of the Study:

• To assess the accuracy of the Dexcom G7 CGM system in critically ill patients within an ICU.
• To compare Dexcom G7 CGM accuracy in the ICU with non-ICU settings and manufacturer's labeling.

Main Methods:

• A prospective, single-center study involving 30 ICU patients with diabetes or stress hyperglycemia requiring insulin therapy.
• Paired sensor glucose values from two Dexcom G7 sensors with blood glucose (BG) measurements obtained via standard ICU care.
• Analysis of accuracy metrics, including mean difference (bias) and mean relative absolute difference (RAD), using plasma and capillary glucose as references.

Main Results:

• A total of 1515 sensor-BG pairs were analyzed, with a mean RAD of 16% (median 12%) against capillary glucose and 13% (median 9%) against plasma glucose.
• The mean RAD in the ICU setting (16%) was slightly higher than outpatient labeling but similar to a non-ICU hospital setting (15%).
• No significant impact of sensor duration or correlation with illness severity on accuracy was observed.

Conclusions:

• The Dexcom G7 CGM demonstrates reasonable accuracy in the ICU, comparable to other hospital settings.
• Further investigation is warranted to determine if CGM can be reliably used for non-adjunctive insulin management in the ICU, including glucose trend monitoring and alarm utilization.