Diabetes is a metabolic disorder that is diagnosed when the fasting and/or postload glucose level rises above well-established thresholds. These thresholds were chosen because they identified people at particularly high risk for retinopathy based on epidemiological data. These data also have shown that people with diabetes and poorly controlled glucose levels have higher risks of retinopathy than people with diabetes and well-controlled glucose levels. 1,2 Moreover, recent studies have shown that the relationship between chronically elevated glucose levels (as measured by A 1c ) and retinal disease is not confined to people with diabetes and is apparent (although less marked) in people with high glucose levels that are below the diabetes cutoffs such as those with impaired glucose tolerance and/or impaired fasting glucose. [3][4][5] Thus, there is a progressive relationship between glycemia and retinopathy that extends below glucose thresholds for diabetes.
Article p 812Clearly, diabetes also is a risk factor for many other serious chronic diseases, including cardiovascular disease. 6 Indeed, a recent meta-analysis of large prospective studies comprising 450 000 people showed that men and women with diabetes are 2 and 3 times more likely, respectively, to die of coronary heart disease than men and women without diabetes. 7 Other studies have shown that the degree of glucose elevation measured by A 1c , fasting glucose, or postload glucose is progressively related to the incidence of cardiovascular outcomes in people with established diabetes and in people without diabetes after adjustment for age and varying numbers of other risk factors. 8 -15 Moreover, several studies that recruited people from both ambulatory and hospitalized settings suggest that there may be a stronger relationship between glycemia and incident cardiovascular outcomes in people without diabetes than in people with diabetes. 8,9,[15][16][17] Such a discrepancy may occur because in people with established diabetes, markers of glycemia are a measure of both adequacy of therapy and exposure to hyperglycemia, whereas they reflect only exposure to hyperglycemia in people without diabetes (in whom glucose and A 1c levels are not targets for therapy).Retinopathy and cardiovascular disease are clearly not the only outcomes related to progressively higher glucose levels. This is perhaps best illustrated by epidemiological analyses of prospective data from the United Kingdom Prospective Diabetes Study (UKPDS), which recruited people with a fasting plasma glucose level Ͼ6 mmol/L 18 and which therefore included individuals with newly diagnosed diabetes as well as some individuals who would be classified as having impaired fasting glucose and not diabetes based on today's diagnostic criteria for diabetes (ie, fasting glucose level Ն7 mmol/L). As noted in the Figure, these analyses showed that progressively higher A 1c levels predicted progressively higher hazards of severe retinal or renal disease, cataracts, myocardial infarction, heart failure, amputa...