956-P: Association Between Glycemia Risk Index (GRI) and Diabetic Retinopathy in Type 2 Diabetes—A Prospective Cohort Study
YAXIN WANG,JINGYI LU,Jiaying Ni,Min Wang,YUN SHEN,Wei Lü,WEI ZHU,YUQIAN BAO,David Rodbard,Robert A. Vigersky,WEIPING JIA,Jian Zhou
DOI: https://doi.org/10.2337/db23-956-p
IF: 7.7
2023-01-01
Diabetes
Abstract:Background: The glycemia risk index (GRI) is a new composite metric for assessing the quality of glycemia based on expert ranking by 330 clinicians of 14-day continuous glucose monitoring (CGM) tracings. Compared with time in range (TIR), GRI has been shown to correlate more closely with the clinicians’ opinions as to the overall quality of glycemia profiles. However, the relationship between GRI and diabetes-related outcomes remains unknown. We aimed to investigate the association between GRI and incident diabetic retinopathy (DR). Methods: A total of 1,204 adult type 2 diabetes without DR at baseline were included between 2005-2019 from a single center in Shanghai, China. GRI was obtained from 3-day CGM data at baseline. Cox proportional hazards regression models were used to estimate the association between GRI and incident DR. Results: During a median follow-up of 8.4 years, 301 patients developed DR. The multivariable-adjusted (age, sex, current smoking status, diabetes duration, BMI, systolic BP, TG, HDL-C, and LDL-C) hazard ratios (HRs) for incident DR across ascending GRI quartiles (≤14 [reference], 15~28, 29~47 and >47) were 1.00, 1.05 (95% CI 0.74-1.48), 1.33 (95% CI 0.96-1.84) and 1.53 (95% CI 1.11-2.11), respectively. For per 1-SD increase in GRI, the risk of DR was increased by 20% (HR=1.20, 95% CI 1.07-1.33) after adjustment for confounders. With respect to TIR, per 1-SD decrease was associated with a 23% (HR=1.23, 95% CI 1.10-1.37) increased risk of DR. There was no statistically significant difference in risk discrimination for incident DR when using GRI and TIR, as measured by C statistic, integrated discrimination improvement (IDI), and net reclassification improvement (NRI) (all P>0.05), after adjustment for confounders. Conclusion: In patients with type 2 diabetes, higher GRI is associated with an increased risk of incident DR, with a predictive value comparable to TIR. The potential value of GRI needs to be explored in future studies. Disclosure Y.Wang: None. R.A.Vigersky: Employee; Medtronic. W.Jia: None. J.Zhou: None. J.Lu: None. J.Ni: None. M.Wang: None. Y.Shen: None. W.Lu: None. W.Zhu: None. Y.Bao: None. D.Rodbard: Consultant; Lilly Diabetes.