AGEs are modifications of proteins or lipids that have become glycated and oxidized following exposure to aldose sugars; they form in vivo in hyperglycemic environments and during aging. AGEs contribute to the pathophysiology of vascular disease in diabetes through accumulation in the vessel walls, where they may perturb cell structure and function. AGEs have also been hypothesized to play a role in atherosclerosis, acute ischemic stroke, and chronic kidney disease (Macsai, 2012; Tang et al, 2013). A number of different therapies to inhibit AGEs are under investigation (Goldin et al, 2006).
Gerrits et al (2008) conducted noninvasive skin autofluorescence (SAF) in 973 type 2 diabetic patients through use of an autofluorescence reader. After a mean follow-up period of 3.1 years, baseline SAF was significantly higher in patients who developed microvascular complications, neuropathy, or (micro)albuminuria, but not in patients who developed retinopathy. This study was the first to observe SAF measurement as an independent predictor of development of microvascular complications in type 2 diabetes.
Hartog et al (2009) investigated whether SAF predicted graft loss following kidney transplantation. They entrolled a total of 302 renal transplant recipients at a median time of 6.1 years post-transplant. They followed the study population for 5.2 years for first occurrence of graft loss. SAF predicted graft loss in a Cox regression multivariable analysis (Hazard ratios 1.83 [1.22-2.75], p = 0.003), adjusted for other identified risk-factors such as patient age, creatinine clearance, protein excretion, high sensitivity C-reactive protein, and human leukocyte antigen-DR mismatching. The investigators concluded that SAF is an independent predictor of graft loss in kidney transplant recipients and that although SAF is not a direct measure of AGEs, the results support a hypothesis that accumulation of AGEs in renal transplant recipients contributes to the development of graft loss.
Smit et al (2010) describe SAF measurement as a noninvasive method of assessing accumulation of AGEs in tissue with low turnover metabolic memory and oxidative stress. One device for measuring AGEs in tissue is the AGE Reader®, which measures tissue accumulation of AGEs by means of fluorescence techniques. It has a light source which illuminates the tissue of interest by exciting fluorescent moieties in the tissue, which will emit light with a different wavelength. In the used wavelength band, the major contribution in fluorescence comes from fluorescent AGEs and therefore the emitted light is detected using a spectrometer. Selective discrimination of specific AGEs can be obtained through use of particular technical adaptations including selection of specific wavelength and modulated or pulsed light sources, so that a more selective discrimination of specific AGEs can be obtained (Diagoptics, 2013).
Skin fluorescence was measured in 105 participants of the Pittsburgh Epdemiology of Diabetes Complications Study of Childhood-Onset type 1 diabetes, who had previously undergone electron beam tomograhy scanning for coronary artery calcification. Study participants’ mean age and diabetes duration were 49 and 40 years, respectively. Measureable coronary artery calcification was found in 71% of participants and univariately cross-sectionally associated with skin fluorescence. However, this association was not maintained after age adjustment. The authors also found that skin fluorescence was both univariately (p < 0.0001) and multivariately ( p = 0.03) associated with coronary artery calcification severity. The authors concluded that the relationship between skin fluorescence and coronary artery calcification appears stronger with more severe calcification, suggesting that skin fluorescence may be a useful marker of coronary artery calcificaiton and coronary artery disease risk and potentially may serve as a potential therapeutic target (Conway, 2010).
A study of 140 type 1 diabetic and 57 nondiabetic subjects was conducted to compare AGE accumulation in the skin of patients in a type 1 diabetic and nondiabetic population. The study also assessed its association with disease duration and metabolic control. The investigators found that mean autofluorescence in the diabetes group was 2.13 ± 0.55, which was significantly higher than in controls (autofluorescence 1.70 ± 0.27, p < 0.05). A significant positive correlation between autofluorescence and patients’ age was found for the whole study population (p < 0.05). A significant positive correlation was also found in diabetic subjects between autofluorescence and diabetes duration (p < 0.05) as well as between autofluorescence and HbA1c levels (p < 0.05). The authors concluded that autofluorescence measurement may be useful as a secondary method of assessing metabolic control as it reflects glycemic control over a longer period of time than that reflected by HbA1c levels (Samborski et al, 2011).
Beisswenger et al (2012) stated that although measurement of SAF has been promoted as a noninvasive technique to measure skin AGEs, the actual products quantified are uncertain. They compared specific SAF measurements with analytically determined AGEs and oxidative biomarkers in skin collagen to determine if these measures are correlated with chronological aging and actinic exposure. SAF was measured at four sites on the arms of 40 non-diabetic subjects. They found poor correlation of AGE-associated fluorescence spectra with AGEs and oxidative products (OPs) in collagen, with only pentosidine correlating with fluorescence at 370(ex)/440(em)nm. Thus, they concluded that SAF measurements at 370(ex)/440(em) nm and 335(ex)/385(em) nm, except for pentosidine, correlate poorly with glycated and oxidatively modified protein in human skin and do not reflect actinic modification. A new fluorescence measurement (440(ex)/529(em) nm) appears to reflect AGEs and OPs in skin.
Hofman et al (2012) noted that AGEs may be involved in aging and development of cardiovascular disease. They further noted that “whether non-invasive measurement of AGE accumulation in the skin may reflect vessel function and vessel protein modification is unknown.” The authors isolated collagen types I and III from the veins of 52 patients by proteolysis to analyze the AGE-modifications in the collagens extracted from residual bypass graft material. The SAF reflected accumulation of AGEs in the body and the pulse wave velocity reflected vessel stiffness. They measured SAF with an autofluorescence reader. They noted that the collagen AGE autofluorescence in vein graft material increased with age and the pepsin digestible collagen fraction was significantly less modified in comparison to the collagenase digestible fraction. Thus, the authors concluded that SAF and pulse wave velocity as non-invasive parameters significantly correlate with the AGE contained in graft material, making them strong predictors of vessel AGE modifications in patients with coronary artery disease. However, the authors also stated that “whether the analysis of the SAF leads to an improvement of the risk stratification in patients suffering from cardiovascular disease has to be further tested.”
Macsai et al (2012) conducted a study to assess whether SAF is influenced by clinical and treatment characteristics in peritoneal dialysis (PD) patients. Their cross-sectional study of 198 PD patients involved utilization of a specific AE Reader device. The authors’ analysis revealed that patients’ age, current diabetes and icodextrine use significantly increased patients’ SAF values (p = 0.015, 0.012, and 0.005, respectively), thus illustrating that in this study group AGE exposure of PD patients with diabetes and on icodextrin solution is increased. The authors noted that further investigation is required to determine whether this finding is due to the icodextrin itself or to a still unspecified clinical characteristic of PD populations treated with icodextrin.
Noordzij et al (2012) evaluated SAFs in patients with carotid artery stenosis with and without coexisting peripheral arery occlusive disease (PAOD) in 56 carotid artery stenosis and 56 age- and sex- matched healthy controls. SAF was found to be higher in patients with carotid artery stenosis compared to the control group (mean 2.81 vs. 2.46, p = 0.002). The authors further noted that patients with carotid artery stenosis and PAOD had an even higher SAF than patients with carotid artery stenosis only (mean 3.29 vs. 2.66, p = 0.003). The investigators concluded that SAF is increased in patients with carotid artery stenosis and PAOD, and that the univariate and multivariate associations of SAF with age, smoking, diabetes, renal insufficiency and PAOD suggest that increased SAF can be seen as an indicator of widespread atherosclerosis.
Current American Association of Clinical Endocrinologists medical guidelines for clinical practice for developing a diabetes mellitus comprehensive care plan do not refer to advanced glycemic endpoints (Handelsman et al, 2011). Although there have been recently published case-control, cross-sectional and case series studies on this topic, the breadth of evidence is such that non-invasive measurement of AGEs in the skin remains experimental and investigational at this time.