Glycemic Threshold for Retinopathy Varies by Race

By Anne Harding

February 28, 2019

NEW YORK (Reuters Health) - Optimal glycemic thresholds for predicting diabetic retinopathy (DR) vary among whites, blacks and Hispanics, new research shows.

"Glycemic control may affect diabetic retinopathy risk differently across race/ethnicity," said study authors Dr. Kevin J. Moore and Dr. Tulay Koru-Sengul of the University of Miami Miller School of Medicine and Dr. Erin C. Dunn of the Floating Hospital for Children at Tufts Medical Center in Boston.

"In addition, the proposed glycemic thresholds demonstrate that controlled diabetes does not rule out diabetic retinopathy risk," they told Reuters Health in a joint email. "We hope that these thresholds emphasize the importance of routine ophthalmologic care among diabetics."

They used 2005-2008 data from the National Health and Nutrition Examination Survey on 1,215 Hispanics, 1,084 non-Hispanic blacks and 3,003 whites, including 661 with evidence of DR on retinal imaging, to identify optimal thresholds for predicting DR.

Optimal predictive hemoglobin A1c threshold was 6% overall, 6% for whites, 6.5% for blacks and 6.4% for Hispanics. "These thresholds might provide health professionals with a more individualized approach to diabetes-related patient care," Dr. Moore and colleagues write in JAMA Ophthalmology, online February 21.

"The 6.5% threshold for black individuals had the highest specificity (85.0%) and sensitivity (52.3%) compared with those for white and Hispanic individuals," the team notes in their report.

Drs. Moore, Koru-Sengul and Dunn told Reuters Health, "Our next steps include looking at diabetic retinopathy and glycemic thresholds prospectively and determining which other sociodemographic variables influence diabetic retinopathy risk."

"A prospective study will allow us to propose more accurate and predictive glycemic thresholds," they said.

SOURCE: https://bit.ly/2UbnXFw

JAMA Ophthalmol 2019.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....