Contributions of pre- and post-prandial blood glucose to glycosylated hemoglobin A1c and their correlation analysis in patients with type 2 diabetes

Abstract

Objective To investigate the contributions of preprandial and postprandial glucose to glycosylated hemoglobin A1c (HbA1c) levels and their correlations in type 2 diabetic patients treated with different drugs. Methods A total of 256 patients with type 2 diabetes (T2DM) with diabetes duration longer than one year and relative fixed treatments were enrolled, which includes 119 males and 137 females.Among them, the average age was (54±14) years, average diabetes duration was 4.5 years, average fasting glucose was 9.6 mmol/L, average HbA1c was 8.4%.The patients were divided into three groups: oral drug treatment group, basal insulin treatment group and premixed insulin treatment group.Each group was then subdivided according to HbA1c levels: <7% group, 7%≤&<8% group, 8%≤&<9% group, ≥9% group.The average of seven-point blood glucose from self-monitored blood glucose (SMBG) from three non-consecutive days was calculated and plotted on glucose-time graph.Meanwhile HbA1c was measured with immunoassay and the correlation between seven-point blood glucose and HbA1c levels was evaluated using Pearson's correlation analysis.On the glucose-time curse, a curve was drawn to connect the three points representing preprandial glucose and the one point representing bedtime glucose.The area between this curve and the horizontal line at 5.6 mmol/L was calculated as the preprandial glucose volume.Another curve (postprandial) was also drawn to connect the three points representing postprandial glucose and the area between preprandial curve and postprandial curve was calculated as postprandial glucose volume.The contribution of pre- and post-prandial glucose to HbA1c was determined by the ratio of pre- and post-prandial glucose volume over total glucose volume (pre- and post-prandial glucose volume together). Results There were significant correlation between HbA1c levels and seven-point self-monitored blood glucose (r=0.38 to 0.76, all P<0.05), and the mean of all seven-point blood glucose had the highest correlation with HbA1c in all treatment groups (r≥0.71, P<0.05). The contribution of pre and post prandial glucose to HbA1c was 59%- 82%, 18%- 41% in oral drug treatment group, 60%- 82%, 18%- 40% in basal insulin treatment group and 38%- 66%, 34%- 62% in pre-mixed insulin treatment group.Tendency analysis showed that the relative contribution of preprandial hyperglycemia increased gradually with increasing values of HbA1c in diabetic patients treated with oral drug and basal insulin and premixed insulin(F=67.46, 168.51, 84.05, P<0.05), but the relative contribution of postprandial hyperglycemia showed a gradual decrease with increasing values of HbA1c.The relative contribution of preprandial hyperglycemia was lower in patients treated with basal insulin than treated with premixed insulin and oral antidiabetic drug(<7% group: 38% vs 66%, 8%≤&<9% group: 63% vs 77%, ≥9% group: 66% vs 82%, χ2=5.94, 5.88, 5.51, all P<0.05). Conclusions There are correlations between HbA1c levels and self-monitored blood glucose testing at various times of the day.Glycemic control and the form of treatment on T2DM patients change the relative contribution of preprandial versus postprandial glucose to HbA1C levels. Key words: Diabetes mellitus, type 2; Hemoglobin A, glycosylated; Hyperglycemia