Endocrine Abstracts

Introduction: Type 2 diabetes is a painless and silent pathology, often overlooked, 30% to 50% underdiagnosed. Its prevalence is constantly increasing. It affects 3 new cases every 10 seconds and more than 6 million per year.

Aim of the work: Compare the HbA1c test to the carbohydrate status established by the HGPO on a sample of an unknown Algerian population with diabetes but at high risk. And try to establish HbA1c thresholds by appreciating the validity of this test for the diagnosis of diabetes mellitus, pre-diabetic states and normal subjects. This test has not yet been validated by Algerian national studies.

Material and method: 500 patients over the age of 40, consultants at the level of primary care structures, volunteers, but at high risk of diabetes, are subjected to a questionnaire, then to screening by the realization of oral hyperglycemia (HGPO), an HbA1c (HPLC method), an FNS, and an ophthalmological examination (fundus). The sensitivity and specificity of HbA1c at different thresholds for the diagnosis of diabetes and pre-diabetes were studied by ROC curve.

Results: 53.2% of patients have dysglycemia: 23.8% have diabetes mellitus, 29.4% have pre-diabetes: 6.8% moderate fasting hyperglycemia (HMJ). 22.6% ITG glucose intolerance. The optimal HbA1c threshold for the diagnosis of diabetes mellitus is 6.27% with a sensitivity of 78% and a specificity of 88%, VPP: 67.88%, VPN: 92.83%, AUC is 0.919. For the diagnosis of pre-diabetes, the optimal HbA1c threshold is 5.83%, with a sensitivity of 71% and a specificity of 81%, VPP: 80.6%, VPN: 72%. The AUC is 0.8157.

Conclusion: The prevalence of diabetes and pre-diabetes in this work is high (23.8% and 29.4%). The use of HbA1c by a standardized method can be a means of screening in subjects at high risk subjects. This HbA1C screening strategy should be verified at the level of the general Algerian population and involve a periodic assessment.