Association of Serum AGR With All-Cause and Cause-Specific Mortality Among Individuals With Diabetes

Abstract Background There is insufficient data to support a link between serum AGR and mortality in individuals with diabetes. This prospective study sought to investigate the relationship between serum AGR and all-cause and cause-specific mortality in adult diabetics. Methods This study included 8508 adults with diabetes from the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2018. Death outcomes were ascertained by linkage to National Death Index records through 31 December 2019. Hazard ratios (HR) and 95% confidence intervals (CIs) for mortality from all causes, cardiovascular disease (CVD), and cancer were estimated using weighted Cox proportional hazards models. Results 2415 all-cause deaths, including 688 cardiovascular deaths and 413 cancer deaths, were recorded over an average of 9.61 years of follow-up. After multivariate adjustment, there was a significant and linear relationship between higher serum AGR levels and reduced all-cause and cause-specific mortality in a dose-response manner. The multivariate-adjusted HR and 95% CI for all-cause mortality (Ptrend<0.0001), cardiovascular mortality (Ptrend<0.001), and cancer mortality (Ptrend<0.01) were 0.51(0.42,0.60), 0.62(0.46,0.83), and 0.57(0.39,0.85), respectively, for individuals in the highest AGR quartile. There was a 73% decreased risk of all-cause death per one-unit rise in natural log-transformed serum AGR, as well as a 60% and 63% decreased risk of mortality from CVD and cancer, respectively (all P<0.001). Both the stratified analysis and the sensitivity analyses revealed the same relationships. Conclusions AGR is a promising biomarker in risk predictions for long-term mortality in diabetic individuals, particularly in those under age 60 and heavy drinker.