Accuracy of the Dexcom G7 Continuous Glucose Monitoring System Following Cardiac Surgery

Monday, March 23, 2026

9:00 AM - 10:00 AM Central Time

Location: Connections Central - RST 01

Mishal Nadeem, MD

Disclosure(s): No relevant financial relationship(s) to disclose.
Mishal Nadeem, MD

Disclosure(s): No relevant financial relationship(s) to disclose.
SI

Steven Insler, MD (he/him/his)

Cleveland Clinic Foundation, Ohio

Disclosure(s): No relevant financial relationship(s) to disclose.
SI

Steven Insler, MD (he/him/his)

Cleveland Clinic Foundation, Ohio

Disclosure(s): No relevant financial relationship(s) to disclose.
AS

Amber Sike, n/a

Research Coordinator
Cleveland Clinic

Disclosure information not submitted.
AS

Amber Sike, n/a

Research Coordinator
Cleveland Clinic

Disclosure information not submitted.
JB

James Bena, MS

Prinicpal Biostatistician
Cleveland Clinic

Disclosure information not submitted.
JB

James Bena, MS

Prinicpal Biostatistician
Cleveland Clinic

Disclosure information not submitted.
ML

M. Cecilia Lansang, MD, MPH

Professor of Medicine, Director of diabetes technology
Endocrinology department, Cleveland Clinic

Disclosure information not submitted.
ML

M. Cecilia Lansang, MD, MPH

Professor of Medicine, Director of diabetes technology
Endocrinology department, Cleveland Clinic

Disclosure information not submitted.

Introduction: Cardiac surgery is associated with hyperglycemia, which can then lead to adverse postsurgical outcomes such as wound infections, acute renal failure, and mortality. Building upon our prior study using the Dexcom G6 Pro, this study seeks to determine if Dexcom G7 continuous glucose monitor (G7) is accurate during the postoperative cardiac surgery period.

Methods: This prospective study received institutional review board approval. G7 was inserted on the upper arm after cardiac surgery. The care team was blinded to the G7 readings and hence not used to make treatment decisions. Clinical (point of care, arterial blood gas, lab glucose) and G7 glucose measures evaluated were restricted to at least 1 hour after a prior clinical measure, and matching G7 data were required within 5 minutes of clinical measures. Patients remained on G7 until discharge or 10 days postoperative.

Results: There were 28 patients, mean age 60.9 +/- 11.3 (SD) years, 71% male, 18% with known diabetes; 85.7% required intravenous insulin infusion. Operative procedures were majority isolated CABG and aortic valve repair/replacement (64%). There were 1,305 clinical glucose readings and 45,035 G7 data points. A total of 707 matched observations were analyzed. Clarke Error Grid Plot showed 79.8% of matched points in zone A, 19.8% zone B, and 0.0% zone C, 0.4% zone D, and 0.0% zone E. There were no significant differences in the rate of Clarke Zone results across differing clinical glucose measurement techniques. The Parkes (Consensus) grid showed similar results: zone A 81.0%, zone B 18.5%, zone C 0.4%, zone D 0.0%, zone E 0.0%. The mean absolute relative difference between clinical and G7 measures was 12.6%. No adverse events were reported.

Conclusions: Almost 100% of G7 and clinical data matching points fell within areas considered as giving clinically correct decisions (zone A) and clinically uncritical decisions (zone B). These findings can assist health care providers in their institutional evaluation of diabetes technology being incorporated into hospital practice.

Acknowledgment: Dexcom provided research support for this project.