A1c's Over Five Percent Increase Risk for Stroke

"In this community-based study population of black and white nondiabetic adults, glycated hemoglobin was superior to fasting glucose for assessment of the long-term risk of subsequent cardiovascular disease, especially at values above 6.0%," write the authors.

The study was led by Dr. Elizabeth Selvin (Johns Hopkins Bloomberg School of Public Health, Baltimore, MD). 

Most of the available data on HbA_1c is from people with diabetes, in whom it's used to monitor glycemic control, Selvin observed. "Until this study, we really haven't had great data on glycated hemoglobin values among people who don't have a diagnosis of diabetes or very long-term follow-up for outcomes such as [incident] diabetes, stroke, coronary heart disease, or mortality."

The current analysis, she said, "demonstrates that HbA_1c is not only an extraordinarily strong predictor of who will develop diabetes, it's also an important marker of cardiovascular risk."

The group's report follows by weeks the publication of Standards of Medical Care in Diabetes -- 2010, an update of diabetes diagnosis and management guidelines from the American Diabetes Association (ADA). The ADA for the first time said that HbA_1c levels >6.5% are sufficient for a diagnosis of diabetes, while levels from 5.7% to 6.4% are a marker of "prediabetes" and indicate increased risk of both incident diabetes and cardiovascular disease.

Of the 14,348 people who attended ARIC planned visit number 2 from 1990 to 1992, which served as baseline in the current analysis, 11,092 were white or African American, nondiabetic, and without a history of cardiovascular disease and yielded blood samples sufficient for measuring HbA_1c. They were followed for development of diabetes and clinical events for a median of about 14 years, Selvin et al reported.

The risks of incident diabetes, coronary heart disease events, and death were significantly increased for all HbA_1c values higher than the reference range of 5.0% to <5.5%; the ischemic stroke risk went up significantly at levels >6.0%. The analysis controlled for patient features and conventional CV risk factors.

Incidence of New Diabetes and Hazard Ratio (95% CI)^A for Diabetes and Other Clinical Outcomes (14-Year Median Follow-Up), by Baseline Glycated Hemoglobin Level, in Nondiabetics:

a. Adjusted for age, sex, race, low-density and high-density cholesterol levels, triglyceride level, body-mass index, waist-to-hip ratio, hypertension, family history of diabetes, education level, alcohol use, physical activity, and smoking status   b. Defined as self-reported diagnosis of diabetes or use of antidiabetic medications   c. Reference for hazard ratios

The analysis hadn't controlled for baseline fasting glucose levels; when it was further adjusted for fasting glucose, the relationship of risk to HbA_1c range didn't change for any of the outcomes.

While the risk of diabetes fell and the CV and stroke risks were unchanged at HbA_1c levels <5.0%, compared with the reference range, all-cause mortality went up significantly at the lower levels.

"The J-shaped relation between the glycated hemoglobin value and the risk of death from any cause suggests that further exploration of the health risks associated with the low-normal glycemic state and possible nonglycemic determinants of glycated hemoglobin is warranted," Selvin et al write.

In a similar analysis by fasting-glucose categories, after adjustment for patient features and conventional CV risk factors, levels of 100 mg/dL to <126 mg/dL and levels >126 mg/dL were significant predictors of sharply increased risk of incident diabetes compared with reference-range levels <100 mg/dL. That didn't change after further adjustment for HbA_1c.

Fasting glucose in the 100 mg/dL to <126 mg/dL range didn't significantly predict coronary heart disease, stroke, or death from any cause, but levels >126 mg/dL were significant predictors of all three end points, in the analysis that adjusted only for patient features and conventional CV risk factors. But even levels >126 mg/dL lost their significance for predicting those three end points after further adjustment for HbA_1c.

Selvin observed that, in clinical practice, assessment of HbA_1c complements rather than replaces fasting glucose, "But our data really do show that a single measure of HbA_1c, compared with a single measure of fasting glucose, provides much more information and is a much more reliable test," she said.

"The new ADA recommendations make clear, and our data really reinforce the idea, that people with an HbA_1c result below 6.5%, below the cut point for new diabetes but still >5.7%, should be targeted for aggressive lifestyle intervention and diabetes prevention and should receive cardiovascular risk-factor screening because -- as we found -- those individuals are at high risk for heart disease and stroke."

1. Selvin E, Steffes MW, Zhu H, et al. Glycated hemoglobin, diabetes, and cardiovascular risk in nondiabetic adults. N Engl J Med 2010; 362:800-811.
2. American Diabetes Association. Standards of medical care in diabetes--2010. Diabetes Care 2010; 33 (suppl 1):S11-S61. Available here. Abstract