Premium
Aging is associated with increased HbA 1c levels, independently of glucose levels and insulin resistance, and also with decreased HbA 1c diagnostic specificity
Author(s) -
Dubowitz N.,
Xue W.,
Long Q.,
Ownby J. G.,
Olson D. E.,
Barb D.,
Rhee M. K.,
Mohan A. V.,
WatsonWilliams P. I.,
Jackson S. L.,
Tomolo A. M.,
Johnson T. M.,
Phillips L. S.
Publication year - 2014
Publication title -
diabetic medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.474
H-Index - 145
eISSN - 1464-5491
pISSN - 0742-3071
DOI - 10.1111/dme.12459
Subject(s) - medicine , prediabetes , postprandial , diabetes mellitus , national health and nutrition examination survey , impaired glucose tolerance , insulin resistance , endocrinology , glucose tolerance test , triglyceride , waist , glycemic , type 2 diabetes , gastroenterology , body mass index , cholesterol , population , environmental health
Aim To determine whether using HbA 1c for screening and management could be confounded by age differences, whether age effects can be explained by unrecognized diabetes and prediabetes, insulin resistance or postprandial hyperglycaemia, and whether the effects of aging have an impact on diagnostic accuracy. Methods We conducted a cross‐sectional analysis in adults without known diabetes in the Screening for Impaired Glucose Tolerance ( SIGT ) study 2005–2008 ( n =1573) and the National Health and Nutrition Examination Survey ( NHANES ) 2005–2006 ( n =1184). Results Both glucose intolerance and HbA 1c levels increased with age. In univariate analyses including all subjects, HbA 1c levels increased by 0.93 mmol/mol (0.085%) per 10 years of age in the SIGT study and by 1.03 mmol/mol (0.094%) per 10 years in the NHANES ; in both datasets, the HbA 1c increase was 0.87 mmol/mol (0.08%) per 10 years in subjects without diabetes, and 0.76 mmol/mol (0.07%) per 10 years in subjects with normal glucose tolerance, all P <0.001. In multivariate analyses of subjects with normal glucose tolerance, the relationship between age and HbA 1c remained significant ( P <0.001) after adjustment for covariates including race, BMI , waist circumference, sagittal abdominal diameter, triglyceride/ HDL ratio, and fasting and 2‐h plasma glucose and other glucose levels, as assessed by an oral glucose tolerance test. In both datasets, the HbA 1c of an 80‐year‐old individual with normal glucose tolerance would be 3.82 mmol/mol (0.35%) greater than that of a 30‐year‐old with normal glucose tolerance, a difference that is clinically significant. Moreover, the specificity of HbA 1c ‐based diagnostic criteria for prediabetes decreased substantially with increasing age ( P <0.0001). Conclusions In two large datasets, using different methods to measure HbA 1c , the association of age with higher HbA 1c levels: was consistent and similar; was both statistically and clinically significant; was unexplained by features of aging; and reduced diagnostic specificity. Age should be taken into consideration when using HbA 1c for the diagnosis and management of diabetes and prediabetes.