Time course of specific AGEs during optimised glycaemic control in type 2 diabetes

Several advanced glycation endproducts (AGEs) are formed in the hyperglycaemic state. Although serum AGEs correlate with average glycaemic control in patients with type 2 diabetes and predict the development of complications, it is not known how serum AGEs change during optimisation of diabetes therapy.

We evaluated the change in serum levels of total AGE and the AGEs CML (Nepsilon-carboxymethyllysine) and MGHI (methylglyoxal-derived hydroimidazolone), as well as markers of endothelial function in 28 subjects with type 2 diabetes, who were poorly controlled on oral agents,before and after the institution of insulin therapy.

Mean subject age (+/- SEM) was 58 +/- 2 years,body mass index 27.7 +/- 0.8 kg/m2, and known duration of diabetes was 8.1 +/- 0.9 years. With insulin treatment fasting blood glucose levels dropped from 12.1 +/- 0.9 mmol/l to 6.9 +/- 0.3 and 8.1 +/- 0.4 mmol/l after three and six months, respectively (both p<0.001), while HbA1c decreased from 10.0 +/- 0.3 to 7.8 +/- 0.2% (p<0.001). Endothelial function improved as indicated by a small but significant decrease in soluble intercellular cell adhesion molecule (sICAM-1) (152 +/- 10 to 143 +/- 8 ng/ml, p<0.02)and sE-selectin (111 +/- 16 to 102 +/-12 ng/ml, p<0.02)levels. In contrast, we observed only a tendency towards a decrease in CML levels (110 +/-22 to 86 +/- 13 microg/mg protein, p=ns), but a small increase of MGHI (from 0.23 +/- 0.02 to 0.29 +/- 0.04 U/mg protein, p<0.02). At baseline, 16 patients were on metformin, which is known to reduce methylglyoxal levels and reduce generation of reactive oxygen species. They had similar levels of CML and MGHI to the 12 non-metformin users, although their HbA1c was lower (9.4 +/- 0.3 vs 10.7 +/- 0.6 %). During insulin, patients receiving concomitant metformin therapy showed a similar course of CML and MGHI to those not taking metformin.

Although insulin therapy improved HbA1c and markers of endothelial function, the levels of serum AGEs did not follow the same time course. This suggests that these specific AGEs are influenced by other factors in addition to overall glycaemia, such as oxidative stress.