1446-P: A1C and Glycated Albumin Combined Improves Detection of Abnormal Glucose Tolerance in Africans

• Published in: Diabetes (New York, N.Y.) Vol. 69; no. Supplement_1
• Main Authors: SUMNER, ANNE E., HOBABAGABO, ARSENE F., SHOUP, ELYSSA M., HORMENU, THOMAS, OSEI-TUTU, NANA H., DUBOSE, CHRISTOPHER, MABUNDO, LILIAN, CHUNG, STEPHANIE T., SACKS, DAVID B.
• Format: Journal Article
• Language: English
• Published: New York American Diabetes Association 01.06.2020
• Subjects: Albumin; Blood glucose; Diabetes; Glucose; Glucose tolerance
• ISSN: 0012-1797; 1939-327X
• DOI: 10.2337/db20-1446-P

In nonwhite populations the prevalence of abnormal glucose tolerance (Abnl-GT) is increasing in the nonobese. However, the prevalence of Abnl-GT in African-born blacks is unknown. As BMI correlates positively with A1C and negatively with glycated albumin (GA) and fructosamine (FA), BMI may influence the diagnostic sensitivity of nonfasting markers of glycemia. Working with Africans, our goals were to determine: a) by BMI category the prevalence of Abnl-GT; and b) the diagnostic value of nonfasting markers of glycemia, individually and combined. In 416 African-born blacks living in America (male: 66%; age: 39±10 (mean±SD), range 20 to 65y, BMI: 27.7±4.5, range 18.2-42.4 kg/m2) OGTT were performed. A1C, GA and FA levels were measured. A1C, GA and FA levels were repeated 11±7 days later in 136 participants. Abnl-GT was defined by glucose: 0h ≥100 and/or 2h ≥140 mg/dL. Thresholds for A1C, GA and FA were defined by the cut-off at the upper quartile for the population (5.7%, 14.2%, 234 µM/L, resp). Analytic tools were κ-statistic, Spearman correlations and the McNemar test. In short, Abnl-GT occurred in 36% (152/416). Of the individuals with Abnl-GT, 68% (104/152) were nonobese. Reproducibility of A1C and GA were excellent (both κ≥0.8) but only moderate for FA (κ=0.6). For this reason, results focus on A1C and GA. BMI positively correlated with A1C (r= 0.18) and negatively with GA (r=-0.18) (both P<0.01). The diagnostic sensitivities of A1C, GA and the combined tests were: 43%, 34% and 61%, resp (P<0.01 compared to A1C alone). By test, 66 people with Abnl-GT were detected by A1C or A1C and GA; whereas 25 people were detected by GA alone. People with Abnl-GT detected by GA only had lower BMI than those detected by A1C (26.8±2.4 vs. 30.4±4, P<0.01). Overall, two-thirds of African-born blacks with Abnl-GT are nonobese. As GA detects Abnl-GT in nonobese individuals not detected by A1C, using both tests could improve screening and lead to earlier intervention. Disclosure A.E. Sumner: None. A.F. Hobabagabo: None. E.M. Shoup: None. T. Hormenu: None. N.H. Osei-Tutu: None. C. DuBose: None. L. Mabundo: None. S.T. Chung: None. D.B. Sacks: Other Relationship; Self; Sebia, Trinity Biotech plc.