LISBON – Skin autofluorescence predicts new-onset type 2 diabetes and cardiovascular disease, according to data from a large prospective study reported at the annual meeting of the European Association for the Study of Diabetes.
Robert van Waateringe, MD, of University Medical Centre Groningen, the Netherlands, presented findings from an analysis of 80,253 individuals participating in the Dutch LifeLines population cohort study showing that skin autofluorescence, a noninvasive biomarker of advanced glycation end products accumulation, was associated with a higher risk of type 2 diabetes (odds ratio, 1.27 per arbitrary unit; 95% confidence interval, 1.06-1.52; P less than .01) after adjustment for common type 2 diabetes risk factors but not after further adjustment for impaired fasting glucose and hemoglobin A1c.
“Skin autofluorescence actually measures the accumulation of advanced glycation end products in the skin, which are also known as AGEs,” Dr. Waateringe explained during an oral presentation.
“AGEs are formed by the nonenzymatic glycation of proteins, lipids, and nucleic acids, which is also known as the Maillard reaction.” There are also exogenous sources, he said, such as heat-processed foods, including fries, steaks, and cookies; AGEs are also found in tobacco smoke.
“The formation and accumulation of AGEs is increased in people with diabetes as a result of hyperglycemia; they are also increased in people with impaired renal function due to their impaired excretion. AGEs cannot only form cross links in the skin but also in the vascular wall, leading to structural and functional tissue impairment,” he added.
It is possible to measure AGEs noninvasively in the skin based on their fluorescent properties, and Dr. Waateringe and his associates used a point-of-care device that has been available for clinical use since 2006 (AGE Reader, Diagnoptics Technologies B.V.). The device has a one-off cost of just under $6,000 (€5,000) and has been “extensively validated” for use by diabetologists and family physicians, with results in a mere 12 seconds, according to the manufacturer. To use the reader, a subject places his or her forearm on top of the unit and allows three separate recordings to be made, which are then averaged. Results are shown according to age, as the accumulation of AGEs increases with advancing age, and can indicate if someone is not at CV risk, is at limited CV risk, is at increased CV risk, or is at definite CV risk.
“We have recently shown that skin autofluorescence is associated with several clinical lifestyle factors in nondiabetic individuals,” Dr. Waateringe said. These factors included older age, being male, having a high BMI and/or high blood glucose levels, the presence of the metabolic syndrome, current smoking, and coffee consumption.
In the current study, Dr. Waateringe and associates included people without a prior history of type 2 diabetes or CVD who underwent baseline investigations between 2007 and 2013 and had a skin autofluorescence measurement available. The development of incident type 2 diabetes was determined between 2014 and 2016 by self-report or by a fasting blood glucose measurement of 6.9 mmol/L or greater or an HbA1c level of 6.5% of more. Self-reports of CVD were obtained, including myocardial infarction, cerebrovascular event, heart failure, or chronic cardiac insufficiency.
As might be expected, the patients who developed incident diabetes (n = 995) versus those who did not (79,252) at 4 years’ follow-up were older (54 years vs. 44 years, P less than .0001), had a larger waist circumference (102 cm vs. 90 cm, P less than .0001), and had a higher fasting glucose (5.9 mmol/L vs. 4.9 mmol/L; P less than .001). They also had higher skin autofluorescence (2.19 vs. 1.91AU, P less than .0001).
Similarly, patients who developed incident CVD (n = 879) versus those who did not (9,368) were older and had a larger waist, higher blood pressure, lower high-density lipoprotein cholesterol, and higher skin autofluorescence.
Skin autofluorescence levels were even higher in people with both new-onset type 2 diabetes and incident CVD than in those with either disease alone, Dr. Waateringe reported.
“Skin autofluorescence predicts type 2 diabetes and cardiovascular disease; however, impaired fasting glucose remained the strongest predictor for type 2 diabetes, and current smoking status was most strongly associated with cardiovascular disease,” he said.
Dr. Waateringe and his coinvestigators noted that skin autofluorescence may be particularly useful in screening for incident type 2 diabetes and CVD in low-risk populations and are looking at using it in combinations with existing risk scores, such as the Finnish Diabetes Risk Score (FINDRISC) and the Systematic Coronary Risk Evaluation (SCORE).
Data so far suggest that combining skin autofluorescence with FINDRISC improves diabetes detection and may prove useful for population screening in which further blood glucose of HbA1c testing could be used to confirm.
Although the present study included people from White European or Caucasian backgrounds, recent studies suggest that this method of detecting AGEs works in people with darker skin tones, although there is a lower reflection and this might influence the readings. Might people getting tanned on holiday also have skewed results? queried one delegate. Dr. Waateringe replied that tans or sunburns were not expected to affect the results.
There is an individual variability in skin autofluorescence of about 6% over the course of the day, Dr. Waateringe acknowledged in response to a question on the technique’s reproducibility.
Dr. van Waateringe reported he had no disclosures. One coinvestigator is founder and shareholder of Diagnoptics Technology, which developed the point-of-care device used in the study.