Abstract
AIMS/HYPOTHESIS: There is controversy regarding the performance of HbA(1c) in old age. We evaluated the prognostic value of HbA(1c) and other glycaemic markers (fructosamine, glycated albumin, fasting glucose) with mortality risk in older adults (66-90 years). METHODS: This was a prospective analysis of 5636 participants (31% with diagnosed diabetes, mean age 76, 58% female, 21% black) in the Atherosclerosis Risk in Communities (ARIC) study, baseline 2011-2013. We used Cox regression to examine associations of glycaemic markers (modelled in categories) with mortality risk, stratified by diagnosed diabetes status. RESULTS: During a median of 6 years of follow-up, 983 deaths occurred. Among older adults with diabetes, 30% had low HbA(1c) (<42 mmol/mol [<6.0%]) and 10% had high HbA(1c) (≥64 mmol/mol [≥8.0%]); low (HR 1.32 [95% CI 1.04, 1.68]) and high (HR 1.86 [95% CI 1.32, 2.62]) HbA(1c) were associated with mortality risk vs HbA(1c) 42-52 mmol/mol (6.0-6.9%) after demographic adjustment. Low fructosamine and glycated albumin were not associated with mortality risk. Both low and high fasting glucose were associated with mortality risk. After further adjustment for lifestyle and clinical risk factors, high HbA(1c) (HR 1.81 [95% CI 1.28, 2.56]), fructosamine (HR 1.96 [95% CI 1.43-2.69]), glycated albumin (HR 1.81 [95% CI 1.33-2.47]) and fasting glucose (HR 1.81 [95% CI 1.24, 2.66]) were associated with mortality risk. Low HbA(1c) and fasting glucose were no longer significantly associated with mortality risk. Among participants without diabetes, associations of glycaemic markers with mortality risk were less robust. CONCLUSIONS/INTERPRETATION: Elevated HbA(1c), fructosamine, glycated albumin and fasting glucose were associated with risk of mortality in older adults with diabetes. Low HbA(1c) and fasting glucose may be markers of poor prognosis but are possibly confounded by health status. Our findings support the clinical use of HbA(1c) in older adults with diabetes. Graphical abstract.