HbA(1c) in the prediction of type 2 diabetes compared with fasting and 2-h post-challenge plasma glucose: The Asturias study (1998-2005)

Abstract

Aim: Fasting plasma glucose (FPG) and the 2-h post-challenge plasma glucose (2hPG) are commonly used to identify those at risk of type 2 diabetes. However, the role of HbA(1c) in this prediction has still not been ascertained.

Methods: The Asturias study is a prospective population-based survey of diabetes and cardiovascular risk factors. Baseline examination, carried out during 1998-1999, involved 1034 individuals, aged 30-75 years, randomly selected to determine the prevalence of type 2 diabetes and prediabetes in the principality of Asturias (northern Spain). In 2004-2005, these same subjects were invited to a follow-up examination, and 700 participated. The present study includes only those who did not have diabetes at baseline. All participants with no known diabetes underwent an OGTT. Baseline HbA(1c) levels were measured by HPLC.

Results: Diabetes had developed in 44 participants at the time of follow-up. Quartiles of baseline HbA(1c) values were 3.4-4.8 (Q1), 4.9-5.1 (Q2), 5.2-5.4 (Q3) and 5.5-6.9 (Q4), and the incidence rates of diabetes by quartiles were 1.0 (0.1-7.1), 4.0 (1.5-10.7), 7.9 (4.0-15.9) and 32.6 (22.9-46.4) cases/1000 person-years, respectively. ROC curve analysis comparing HbA(1c), FPG and 2hPG in the prediction of diabetes showed areas under the curve (ROC-AUC) of 0.80 (0.74-0.86), 0.83 (0.77-0.90) and 0.79 (0.72-0.87), respectively. The combination of FPG and HbA(1c) had the best predictive performance with an ROC-AUC of 0.88 (0.82-0.93).

Conclusion: Our study indicates that HbA(1c) is strongly predictive of new-onset diabetes in this northern Spanish population, and was similar to FPG and 2hPG in predictive capability. Also, the combined measurement of FPG and HbA(1c) improved their individual predictive performance.