The incongruous finding, among patients with impaired glucose tolerance (IGT) or impaired fasting glucose (IFG) who participated in the landmark Diabetes Prevention Program (DPP), suggests that current cutoffs used to diagnose diabetes may need to be revised, said Dr. Hamman, professor and chair of the department of preventive medicine and biometrics at the University of Colorado, Denver.

More patients must be studied to determine whether the retinal lesions found in these “prediabetic” individuals represent early diabetic changes or perhaps are more indicative of arteriolar changes in people at risk for atherosclerotic vascular disease in general. Still, the finding does suggest that “a good retinal exam during the transition from prediabetes to early diabetes is important,” Dr. Hamman said at a press briefing.

Little is known about exactly when retinopathy develops in patients with type 2 diabetes, because the diagnosis often lags years behind the onset of high blood sugar. The DPP, which followed 3,234 high-risk individuals at 6-month intervals, afforded a unique opportunity to date precisely the development of retinopathy in relation to diabetes onset during the trial and subsequent long-term follow-up, he noted.

At a mean of 5.5 years between randomization in DPP and the taking of retinal photos, retinopathy of any degree was found in 15.4% of the 301 who developed diabetes during the trial and in 9.6% of 585 who did not. Diabetic retinopathy of grade 20 or higher—considered more indicative of true diabetes-related changes—was found in 12.5% of those with diabetes of short duration and in 7.6% of those who remained as IGT or IFG.

Similarly, the proportions who had only microaneurysms were 10.6% and 6.6%, respectively. Moderate nonproliferative diabetic retinopathy was detected in 1.9% of those who did not meet the criteria for diabetes during the trial and in 1.0% of those who did, a nonsignificant difference that nonetheless follows the same trend. Another nonsignificant trend was the increase in the prevalence of diabetic retinopathy by duration of diabetes, from 8.9% of those with diabetes diagnosed within 1 year to 14.5% among the subjects diagnosed more than 4 years earlier.

Thus, aside from the prediabetes findings, these data also show that retinopathy may appear far earlier in the course of diabetes than was previously thought. “Just 6–12 months after diabetes onset, almost 13% had retinopathy,” Dr. Hamman said.
Among subjects who did not develop diabetes, triglycerides were the only other risk factor associated with development of retinopathy, with no differences in other lipid parameters, hemoglobin A1c, or blood pressure between those who developed retinopathy and those who did not.

Among those who developed diabetes, blood pressure and HbA1c levels were associated with retinopathy. Prior to their diabetes diagnosis, those with retinopathy had a mean blood pressure of 129/80 mm Hg, compared with 124/78 mm Hg among those without retinopathy, a significant difference. After the diabetes diagnosis, those values were 127/79 mm Hg and 123/77 mm Hg, respectively; here, only the systolic pressure difference is significant.

Prior to the diagnosis of diabetes, those with retinopathy had a mean HbA1c of 6.60% and those without, 6.26%. After diagnosis, those A1c values were 6.52% and 6.32%, a highly significant difference.

Although the numbers are small, the prevalence of diabetic retinopathy in high-risk individuals who ultimately developed diabetes and in those who stayed below the cutoff are far higher than the 3% rate of retinopathy in the general population.

“So, there is definitely something going on,” said Dr. Hamman, adding that the study in progress—which is funded by the National Institutes of Health and the Centers for Disease Control and Prevention—plans to analyze a larger number of subjects, and to investigate whether prevention of diabetes can also prevent the development of retinopathy.