Evaluation of HbA1c criteria for diagnosis of diabetes mellitus: a retrospective study of 12 785 type 2 Saudi male patients

Recently, American Diabetic Association has recommended glycated hemoglobin (HbA1c ≥6.5%) as an alternate to fasting plasma glucose (FPG ≥7.0 mmol/L) for diagnosis of diabetes. However, studies from different groups showed inconsistent results with the use of HbA1c criteria. We examined the validity of HbA1c cut-point of 6.5% for diagnosis of diabetes. A total of 12 785 male diabetic patients (FGP ≥7.0 mmol/L), aged 56.27 ± 13.32 years were included. The average values of FPG and HbA1c of all the 12 785 patients were 10.127 ± 0.026 mmol/L and 8.729 ± 0.013%, respectively. Sub-grouping of patients into different age categories showed significantly high levels of FPG (10.934 ± 0.123 mmol/L) in the youngest group (age, ≥20-35 years) as compared to FPG (ranged from 10.021 ± 0.052 to 10.190 ± 0.050 mmol/L) in patients with other age categories. The level of HbA1c was highest in the youngest group (8.809 ± 0.056%) and lowest in the oldest group (8.653 ± 0.082%). There was a significant correlation between FPG and HbA1c (R = 0.571, p < 0.001). There were 484 patients below the diagnostic threshold (HbA1c <6.5%), resulting in 3.78% false negative predictions. Majority of the false negative patients were in the age group of 40-75 years and had borderline FPG (7.0-8.0 mmol/L) and HbA1c (6.0-6.5%). These findings suggest that Saudi individuals with HbA1c between 6.0% and 6.5% may be considered as "probable diabetic" and their status should be verified by combined FPG and HbA1c criteria.