Prediction of cardiovascular events by skin auto-fluorescence: the DIABAGE follow-up study.

AIM: Advanced glycation end-products (AGEs) are known to play a role in the pathophysiology of type 1 diabetes (T1D) complications. The aim of this study was to assess the predictive value of AGEs indirectly evaluated by skin auto-fluorescence (SAF) on the occurrence of cardiovascular events (CVEs) in T1D.

METHODS

We measured baseline SAF in T1D patients with at least 10 years history of diabetes and assessed incident CVEs. An optimum threshold of SAF was determined using ROC curve, and its predictive value was assessed by Cox proportional regression.

RESULTS

The study included 179 patients, 53 % of whom were women. At baseline, the mean age was 47.7 ± 15.9 years, the mean duration of diabetes was 26.4 ± 12.2 years. Median HbA1c was 7.7 % (7.3-8.7) and median LDL cholesterol was 2.58 mmol/l (2.14-3.07). Median follow-up was 7.4 years (6.85 - 7.7) with 34 CVEs in 24 patients. The median SAF level was 2.7 (2.3-3.1) in patients with incident CVEs and 2.1 (1.8-2.6) in patients without CVEs. The optimum threshold of SAF to differentiate patients with or without incident CVEs was 2.2. The occurrence of CVE was predicted by the optimum SAF threshold in the unadjusted model (HR 6.46), but also after adjustment with different models (HR 3.15-5.05).

CONCLUSION

SAF level is higher in people living with T1D who will present CVEs. Furthermore, SAF threshold of 2.2 predicted the occurrence of CVE. If these results are confirmed, SAF could be a useful marker in cardiovascular risk stratification in T1D.