Continuous Glucose Monitoring Reveals Associations of Glucose Levels with QT Interval Length

Background: QTc interval lengthening during hypoglycemia is discussed as a mechanism linked to sudden death in diabetes patients and the so-called "dead in bed syndrome.'' Previous research reported a high inter-individual variability in the glucose-QTc association. The present study aimed at deriving parameters for direction and strength of the glucose-QTc association on the patient level using combined Holter electrocardiogram (ECG) and continuous glucose monitoring. Methods: Twenty type 1 diabetes patients were studied: mean (SD, range) age, 43.6 (10.8, 22-65) years; gender male (n [%]), 10 (50.0%); mean (SD) hemoglobin A1C, 8.5% (1.0%); and impaired hypoglycemia awareness (n [%]), six (30.0%). Continuous interstitial glucose monitoring and Holter ECG monitoring were performed for 48 h. Hierarchical (mixed) regression modeling was used to account for the structure of the data. Results: Glucose levels during nighttime were negatively associated with QTc interval length if the data structure was accounted for (b [SE] = -0.76 [0.17], P = 0.000). Exploratory regression analysis revealed hypoglycemia awareness as the only predictor of the individual strength of the glucose-QTc association, with the impaired awareness group showing less evidence for an association of low glucose with QTc lengthening. Conclusions: Mixed regression allows for deriving parameters for the glucose-QTc association on the patient level. Consistent with previous studies, we found a large interindividual variability in the glucose-QTc association. The finding on impaired hypoglycemia awareness patients has to be interpreted with caution but provides some support for the role of sympathetic activation for the QTc-glucose link.