Just like any word in the dictionary, diabetes has a specific definition. But instead of Webster’s calling the shots, medical groups like the American Diabetes Association (ADA) and the World Health Organization decide what criteria are used to diagnose diabetes.

Currently, people are considered to have the disease if their fasting plasma glucose (FPG) is 126 mg/dl or higher, as determined by a blood test. That threshold for diagnosis comes from studies of the diabetic eye disease retinopathy, which found a marked increase in incidence at and above the 126 mg/dl level.

It turns out, however, that those studies may have had limitations, says Tien Wong, MD, professor of ophthalmology at the University of Melbourne. Wong and his colleagues went looking for an FPG threshold for retinopathy in three studies of 11,000 people, a larger and more diverse group than the populations in the older studies. They worked with the current technological gold standard—multiple-view retinal photographs—for retinopathy screening, a tool that earlier researchers did not use.

What they found was a dramatic departure from the earlier studies: The relationship between retinopathy and FPG started at FPG levels well below the current threshold for the diagnosis of diabetes.

Retinopathy rates still rise as FPG increases, but there is no distinct cut-off point between low and high risk, as was suggested by the earlier research, says Wong. “This really demonstrates that the impact of glucose on the blood vessels does not respect a particular threshold,” he says.

Other research has also hinted that the risk of retinopathy increases as one inches toward an FPG of 126 mg/dl, but according to Wong, this is the first study that has used the right tools and study group to adequately define the retinopathy problem.

Wong and his team will next search for a measure they hope will provide a more appropriate cut-off point for a diabetes diagnosis. It should be noted, however, that most of the retinopathy noted in people with FPG levels that were non-diabetic was mild in nature. In addition, their retinopathy may have been not the direct result of increased FPG levels, but instead due to high blood pressure and other vascular diseases.

For now, the current guidelines hold, but the ADA is convening a group of diabetes experts to consider whether and how they should be revised.

Wong’s study was published in the Mar. 1, 2008, issue of The Lancet.