Using banked urines from 156 DCCT participants of the former skin biopsy cohort (Diabetes 48:870,1999) we tested whether free AGEs measured at DCCT year 4 (Timepoint TP1), year 8(TP2) and EDIC year 17 (TP3) are associated with risk of microvascular complications progression.

Methods: Fructose-lysine(FL), glucosepane(GSPN), carboxymethyl-lysine(CML), methylglyoxal (MG-H1) and 10 other markers were measured by LC/MS/MS. Data were analyzed using correlation coefficients, while associations with risk of complications using Cox proportional hazards (PH) models without adjustment for multiplicity.

Results: FL (P<0.005) and GSPN (P<0.004) were significantly lower in the intensive therapy group at TP2. At all 3 TPs, mean A1C correlated with FL (r=0.31) and GSPN (r=0.30) (P<0.0001). In unadjusted Cox PH models, higher GSPN levels were associated with sustained eGFR<60 (P=0.007), cardiac autonomic neuropathy (CAN) (P=0.008) and borderline with confirmed clinical neuropathy (CCN) (P=0.05, NS). FL and CML were associated with sustained microalbuminuria (SA) both before (P≤0.001, 0.012) and after (P≤0.034) adjustment for A1C. Upon adjustment for A1c, age and gender using all 3 TPs, methionine sulfoxide/MetSOX (P=0.001), 3-nitrotyrosine (P=0.003), and FL and G-H1 (P<0.027) were associated with PDR; pentosidine (P=0.041) with clinically significant macular edema; MetSOX (P=0.013) and GSPN (P<0.034) with sustained eGFR<60; FL (P=0.015), CML (P=0.028) and MG-H1 (P=0.042) with SA. LW-1/glucuronidine was associated with CCN both before (P<0.001) and after adjustment for A1C (P<0.001). Chi square statistics were larger than A1c for GSPN with CAN, MetSOX with sustained eGFR<60 and LW-1/glucuronidine for CCN, suggesting stronger associations.

Conclusions: FL, GSPN, and CML in urine consistently emerge as the AGEs most significantly associated with risk of subsequent microvascular complications.


D. Sell: None. J. Lachin: Board Member; Self; Tolerion, Inc. I. Bebu: None. V.M. Monnier: Consultant; Self; Juvenescence Ltd.


National Institute of Diabetes and Digestive and Kidney Diseases (DP3DK101123)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at