The Association of Time-in-Range and Time-in-Tight-Range with Retinopathy Progression in the Virtual Diabetes Control and Complications Trial Continuous Glucose Monitoring Dataset

Background: In a prior work, a virtual continuous glucose monitoring (CGM) trace was generated for each of the 1441 participants in the landmark Diabetes Control and Complications trial (DCCT). These new data allow us to compare whether time-in-tight-range (TITR) is a better predictor of diabetic microvascular complications (specifically retinopathy development or progression) than time-in-range (TIR). Methods: Discrete Cox proportional hazard models were used to calculate the hazard ratios (HRs) for the development/progression of retinopathy. Results: For a 1.0 standard deviation (SD) change, the adjusted HR (95% confidence interval) was 2.67 (2.33-3.06) for TIR, 2.74 (2.36-3.18) for TITR, and 2.37 (2.13-2.65) for HbA1c; a similar pattern of results was obtained for a 0.5 SD change. Computing Harrell's C-statistic showed that a survival model adjusted for TIR, TITR, or HbA1c had similar predictive performance. Conclusion: The associations of TIR and TITR with retinopathy development or progression were similar to HbA1c in the virtual DCCT CGM dataset.