Comment on Koska et al. Advanced Glycation End Products Predict Loss of Renal Function and High-Risk Chronic Kidney Disease in Type 2 Diabetes. Diabetes Care 2022;44:684–691

We were interested by the recent article from Koska et al. (1), who related high levels of circulating advanced glycation end products (AGEs) to later loss of renal function in participants in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial and the Veterans Affairs Diabetes Trial (VADT). The fluorescent properties of some AGEs allow an indirect evaluation of their concentrations in tissues by measuring the skin autofluorescence (SAF) (2). In type 1 diabetes, we have shown that the SAF could predict later renal impairment, defined as an estimated glomerular filtration rate <60 mL/min/1.73 m² (3). Fourteen years ago, the Zwolle Outpatient Diabetes Project Integrating Available Care (ZODIAC) study had shown that the SAF predicted later microalbuminuria in well-controlled type 2 diabetes of short duration (4). More recently, Jin et al. (5) reported that the SAF predicted the progression of kidney disease in 4,030 subjects from the Hong Kong Diabetes Biobank. It is, however, unknown whether high circulating levels of accumulated AGEs predict more severe renal outcomes, like end-stage renal disease (ESRD) or doubling of serum creatinine.

We have followed 519 subjects hospitalized in our unit for uncontrolled or complicated type 2 diabetes to determine whether their SAF was related to later renal end points: doubling of serum creatinine and ESRD. The subjects were 62 ± 9 years old, mainly men (58%), with a duration of diabetes of 14 ± 10 years that was poorly controlled (HbA_1c 8.7% ± 1.8%), and 232 (44.7%) had a history of diabetic kidney disease (DKD). During the 53 (interquartile range 32–73) months of follow-up, 17 subjects developed ESRD (16 started on dialysis and 1 had a renal transplantation), and 14 others doubled their serum creatinine.

The 31 subjects who experienced such a renal event were 2 years older (not significant) and had 4 years longer diabetes duration (P = 0.038), and most of them had a history of DKD (83.9% vs. 16.1%, P < 0.001). They did not differ for sex and HbA_1c, but their SAF levels were higher: 3.0 ± 0.7 vs. 2.6 ± 0.6 arbitrary units (P < 0.001). By Cox regression analysis adjusted for age, sex, duration of diabetes, HbA_1c, and DKD, the SAF was related to later renal events (for >1 arbitrary units: hazard ratio 1.85, 95% CI 1.13–3.05).

There are now several pieces of evidence showing that AGEs contribute to all stages of DKD, from microalbuminuria (4) to macroproteinuria (1), loss of renal function (1,5), and finally ESRD, in both types of diabetes (3). We trust that the SAF is of special interest because it reflects AGEs accumulated in tissues where they exert their deleterious effects, and it is a simple and noninvasive measurement. However, circulating AGEs as measured by Koska et al. (1) may be more modifiable in response to therapeutic interventions. We wonder whether the authors had the opportunity to perform a second measurement in some of the participants in their cohorts or if they have any information on the possible reversion of high AGE scores.