A Modified CGM Algorithm Enhances Data Availability While Retaining iCGM Performance

Area of Science:

• Biomedical Engineering
• Medical Device Technology
• Diabetes Technology

Background:

• Continuous glucose monitoring (CGM) is crucial for diabetes management.
• Sensor errors in CGM systems can disrupt data collection and impact patient care.
• The Dexcom G6 CGM System previously experienced sensor errors that affected data availability.

Purpose of the Study:

• To evaluate an enhanced algorithm for the Dexcom G6 CGM System.
• To assess the impact of the algorithm modification on sensor error frequency and duration.
• To determine if accuracy was maintained after the algorithm update.

Main Methods:

• Post-processing of raw signals from Dexcom G6 pivotal trials (NCT02880267).
• Assessment of data availability changes after a limited real-world launch.
• Comparison of accuracy metrics (e.g., %20/20, MARD) before and after algorithm modification.

Main Results:

• Overall accuracy (%20/20) remained comparable (91.7% vs. 91.8%), and MARD was unchanged.
• Mean data gap due to sensor error was nearly halved.
• Total time spent in sensor error decreased by 59% in pivotal trials and 43% in a real-world launch.

Conclusions:

• The enhanced Dexcom G6 algorithm successfully reduced sensor error frequency and duration.
• Improved data availability from the CGM system can enhance patient experience.
• Increased data availability facilitates better integration with automated insulin delivery systems.