[Correlation between glycemic variability and glycosylated hemoglobin level during follow-up in elderly male patients with type 2 diabetes in Beijing].

ABSTRACT

Objective: To investigate the relationship between glycemic variability and glycosylated hemoglobin (HbA1c) level during follow-up in elderly male patients with type 2 diabetes. Methods: Retrospective cohort study. A total of 200 elderly male patients who received continuous glucose monitoring from January 2007 to January 2011 were recruited in the Second Medical Center of PLA General Hospital. The subjects were divided into two groups according to baseline mean amplitude of glycaemic excursion (MAGE) level, including MAGE <3.9 mmol/L group (n=114) and MAGE ≥3.9 mmol/L group (n=86). The correlation between baseline MAGE and mean HbA1c during follow-up were evaluated by univariate Pearson correlation analysis and multivariate linear regression analysis. Results: Baseline characteristics including age, body mass index, waist circumference, smoking, drinking, fasting blood glucose, blood lipid and blood pressure were comparable between MAGE <3.9 mmol/L group and MAGE ≥3.9 mmol/L group. The average follow-up period was 12.5 years. The mean HbA1c during follow-up in MAGE ≥3.9 mmol/L group was significantly higher than that in MAGE <3.9 mmol/L group (7.23%±0.72% vs. 6.91%±0.77%, t=-2.94, P=0.004). The proportion of mean HbA1c <7.0% during follow-up in MAGE ≥3.9 mmol/L group was 44.2% (38/86), which was significantly lower than that in MAGE <3.9 mmol/L group [60.5% (69/114), χ2=5.26, P=0.022]. In univariate analysis, MAGE at baseline was correlated with the mean HbA1c during follow-up (r=0.306, P<0.001). Multivariate linear regression analysis suggested that the baseline MAGE remained an independent influential factor of mean HbA1c (ß=0.09, 95%CI 0.03 to 0.15, P=0.006, R2=0.31) after several confounding factors were adjusted. Conclusions: With the increased glycemic variability at baseline, mean HbA1c level during follow-up is accordingly elevated. The glycemic variability at baseline is independently related to mean HbA1c level during follow-up in elderly male patients with type 2 diabetes.