Introduction: Older adults with diabetes have variable prognosis. There is critical need to improve risk stratification among this population to understand who is most likely to experience adverse outcomes. Low 1,5-anhydroglucitol (1,5-AG) is a biomarker of glycemic variability and has demonstrated value for identification of middle-aged adults with diabetes at risk for major clinical outcomes. Total hospitalizations are a useful summary measure of poor health outcomes. It is unknown whether 1,5-AG can identify older adults at risk for hospitalizations and all-cause mortality.
Methods: We included 2,061 participants from the Atherosclerosis Risk in Communities (ARIC) Study with diagnosed diabetes who attended the 2011-2013 visit. We dichotomized 1,5-AG (≥6μg/mL; <6μg/mL) and followed participants until December 31, 2015. We examined the associations of 1,5-AG with total and diabetes-related hospitalizations using negative binomial regression and all-cause mortality using Cox regression.
Results: Participants ranged in age from 67-90 years, 57% were female, 30% were black, and 17% had 1,5-AG <6μg/mL. Median HbA1c was 6.2% in those with 1,5-AG ≥6μg/mL and 7.8% in persons with 1,5-AG <6μg/mL. During a median of 3.6 years of follow-up, there were 2,813 hospitalizations (1,689 diabetes-related) and 247 deaths. Compared to 1,5-AG ≥6μg/mL, individuals with 1,5-AG <6μg/mL had a significantly higher risk of hospitalizations, diabetes-related hospitalizations, and death (Table). After adjustment for diabetes medication use or HbA1c, associations with hospitalizations were attenuated and non-significant, while the relationship with all-cause mortality remained.
Conclusion: Among older adults with diagnosed diabetes, glycemic variability may be an important risk factor for major short-term complications.