**8. The** *'Metabolic Tipping-Point'* **and Its Effect on Intervention**

There is a large body of observational and experimental research dating from the 1950s showing that although body mass and composition and concomitant basal energy expenditure are the major determinants of caloric consumption [60–64], PA is the major *modifiable* determinant of consumption, expenditure, and storage [16,27,28,31,34–37,65–68]. Thus, because PA plays an essential role in all aspects of metabolism, we previously coined the term *'Metabolic Tipping-point'* to denote the amount of PA necessary to prevent overconsumption and weight gain [1,12]. As briefly explained below, and in detail elsewhere [1,12], this concept offers a concise framework for understanding the heterogeneity of response of caloric consumption and body and fat mass to altered levels of PA.

As depicted in Figure 1, PA, body mass, and caloric consumption have complex, nonlinear relations [16]. When an individual's PA declines below their lower metabolic tipping-point (the left side of Figure 1), caloric intake declines more slowly than energy expenditure (a nonlinear relation). This leads to increments in body fat and mass, and decrements in skeletal muscle insulin-sensitivity. If habitual, these individuals will develop acquired obesity and T2DM—dependent on fat-cell plasticity and pancreatic beta-cell function. Nevertheless, any intervention that increases their PA above their lower metabolic tipping-point will reduce hyperphagia, positive energy balance, and prevent further gains in body and fat mass. Nevertheless, as explained in a previous section, *caloric restriction is essential* if the excess body mass is to be reduced because interventions that rely exclusively on PA and exercise have trivial effects on body mass.

**Figure 1.** Relations between PA, Body Mass, and Energy Intake (adapted from [69]). As PA declines below the lower metabolic tipping-point into the 'Sedentary' range (left panel), energy intake and energy expenditure become dissociated due to insufficient PA. Body mass begins to increase as energy balance becomes positive and insulin sensitivity is diminished.

When individuals maintain PA levels between the upper and lower metabolic tippingpoints (the center portion of Figure 1), their body and fat mass remain stable, regardless of increments and decrements in PA within this range. This occurs because of a linear relation between caloric consumption and expenditure at moderate levels of PA. Thus, as

PA increases, caloric consumption increases in parallel. To be precise, the *nonlinear* relations below the lower tipping-point explains why decrements in PA lead to increments in body and fat mass in highly sedentary individuals, whereas the *linear* relation between PA and consumption in the range between the upper and lower metabolic tipping-points explains why increased PA and exercise have little or no effect on body mass in individuals who are already moderately active.

Conversely, when individuals increase their PA above their upper metabolic tippingpoint (the right side of Figure 1), they experience declines in caloric consumption, basal energy metabolism, energy expenditure, and body and lean mass. This level of PA is not sustainable and leads to incomplete recovery, reduced physical performance, injury, and exhaustion [70].

In summary, the left side of Figure 1 depicts the *nonlinear* relations between caloric consumption and expenditure, and the concomitant development of acquired obesity and T2DM. The center panel depicts the *linear* relations between PA and caloric consumption and explains why exercise interventions without caloric restriction will not reduce body and fat mass. The right side of Figure 1 depicts unsustainable levels of PA that lead to the loss of body and lean mass. Therefore, it is the transition from a nonlinear to a linear relation between caloric consumption and expenditure as PA increases from a sedentary to an active lifestyle that explains the heterogeneity of response to diet and exercise in individuals with varied levels of baseline PA.

Nevertheless, what Figure 1 does not depict is how an individual's obese subtype impacts their metabolic tipping-points. Because the acquired obese phenotype is essentially a behavioral phenomenon, any intervention that increases PA and reduces caloric consumption will be successful in the early and mid-stages of phenotypic development; however, the longer the physical inactivity-induced metabolic perturbations continue, the less amenable to intervention the acquired subtype becomes. In this respect, long-standing acquired obesity will mimic the inherited subtype in its response.

Conversely, individuals with an inherited subtype represent a continuum of irreversible structural (anatomic) and physiologic (metabolic) perturbations that are inherent to their phenotype. As such, the amount of PA and caloric restriction necessary to reduce body mass and maintain weight loss depends on where they fall in the continuum of perturbations—from mild to extreme. The more extreme an individual's inherited obese phenotype, the higher their metabolic tipping points, and the greater the amount of PA and caloric restriction required to prevent overconsumption and achieve and maintain a healthy weight.

Nevertheless, the physical and psychological burdens induced by large amounts of PA and severe caloric restriction are beyond the perseverative capacity of most humans. As such, the long-term maintenance of weight loss becomes an increasingly unachievable goal as the structural and physiologic perturbations become more severe. Therefore, the management objective for individuals with inherited subtypes should be along the continuum of 'fit but fat'. The refusal to appreciate this reality has led to unrealistic expectations, management 'failure', and the stigmatization of individuals with an inherited obese phenotype [71,72].

#### **9. Assumptions and Limitations**

Our 'perspective' is based on several assumptions that may limit our conclusions. The most critical is that obesity and cardiometabolic diseases are wholly anatomical (structural) and physiological (metabolic) disorders. Thus, we posit that if psychological, social, economic, or other non-physiologic phenomena influence obese or diabetic phenotypes, they must act through cellular mechanisms that cause increments in skeletal muscle-cell insulin-resistance and its sequelae (e.g., hyperphagia, adipogenic caloric partitioning, and increased fat-cell mass and number).

Although a large body of experimental evidence demonstrating the causal effects of PI on skeletal muscle-cell insulin resistance and its sequelae exists, the only support for speculations regarding the effects of psychological, social, and economic phenomena is correlational.

Moreover, we assert that distinguishing between etiology and treatment is critical for discussions revolving around the roles of PA, genes, diet, and exercise. For example, we contend that although specific macro-nutrients are not causal to obesity and other disease states, except as a source of calories (for details please see [69,73–76]), we argue strongly that caloric restriction with further reductions in carbohydrates are essential protocols for reducing body and fat mass and the treatment of acquired obesity and T2DM.

Finally, although our work on nongenetic inheritance and the developmental origins of disease is rigorous, consilient, and supported by voluminous research across species (please see [12] for details), our theories are novel and may therefore appear controversial to those unfamiliar with this emerging area of research and science (for reviews see [3,17–19]. Nevertheless, it remains to be seen if our conclusions withstand the 'test of time'.

#### **10. Summary and Conclusions**

The age and rate at which an individual's obese phenotype develops is a strong determinant of its response to intervention. Thus, the development of effective management strategies necessitates a personalized approach that incorporates the subtyping of obese phenotypes by etiologic status (acquired or inherited). The acquired phenotype denotes the development of excessive adiposity after puberty and is essentially a behavioral phenomenon induced by low levels of PA and concomitant hyperphagia (overconsumption). Thus, effective prevention and treatment strategies can be based on diet and exercise [32,47,77]. Although this subtype is amenable to lifestyle interventions in the early stages of development, the longer the PI, excessive SB, and overconsumption continue, the less amenable to intervention this subtype becomes.

In contrast, inherited obesity subsumes all forms of excessive adiposity that develop prior to pubescence (pediatric and childhood). The prevention of non-genetic inherited obese phenotypes in the next generation necessitates adequate levels of PA by the current generation of young females, potential mothers, and pregnant women. Nevertheless, once instantiated during the prenatal period, this subtype has irreversible structural (anatomic) and physiologic (metabolic) perturbations that are not amenable to intervention because no amount of diet and exercise can reduce the excessive number of fat cells and adipogenic partitioning, or significantly improve skeletal muscle function. Therefore, the objective in the management of inherited subtypes is the development of a 'fit but fat' phenotype. Importantly, because the amount of PA and caloric restriction necessary for the maintenance of weight loss with an inherited subtype may be beyond the physical and psychological capabilities of most individuals, it should not be the goal.

In closing, clinicians and investigators must recognize that despite its ubiquity, obesity is not a homogenous condition. Moreover, because obesity is a complex and idiosyncratic phenotype determined by inherited, behavioral, and environmental factors, a personalized approach based on etiologic subtype is essential for successful health management.

**Author Contributions:** Conceptualization, E.A. and C.J.L.; writing—original draft preparation, E.A.; writing—review and editing, E.A. and C.J.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Is There a Link between Obesity Indices and Skin Autofluorescence? A Response from the ILERVAS Project**

**Enric Sánchez 1,†, Marta Sánchez 1,†, Carolina López-Cano 1, Marcelino Bermúdez-López 2, José Manuel Valdivielso 2, Cristina Farràs-Sallés 3, Reinald Pamplona 4, Gerard Torres 5,6, Dídac Mauricio 7,8, Eva Castro 2, Elvira Fernández <sup>2</sup> and Albert Lecube 1,8,\***


**Abstract:** There is controversial information about the accumulation of advanced glycation endproducts (AGEs) in obesity. We assessed the impact of total and abdominal adiposity on AGE levels via a cross-sectional investigation with 4254 middle-aged subjects from the ILERVAS project. Skin autofluorescence (SAF), a non-invasive assessment of subcutaneous AGEs, was measured. Total adiposity indices (BMI and Clínica Universidad de Navarra-Body Adiposity Estimator (CUN-BAE)) and abdominal adiposity (waist circumference and body roundness index (BRI)) were assessed. Lean mass was estimated using the Hume index. The area under the receiver operating characteristic (ROC) curve was evaluated for each index. Different cardiovascular risk factors (smoking, prediabetes, hypertension and dyslipidemia) were evaluated. In the study population, 26.2% showed elevated SAF values. No differences in total body fat, visceral adiposity and lean body mass were detected between patients with normal and high SAF values. SAF levels showed a very slight but positive correlation with total body fat percentage (estimated by the CUN-BAE formula) and abdominal adiposity (estimated by the BRI). However, none of them had sufficient power to identify patients with high SAF levels (area under the ROC curve <0.52 in all cases). Finally, a progressive increase in SAF levels was observed in parallel with cardiovascular risk factors in the entire population and when patients with normal weight, overweight and obesity were evaluated separately. In conclusion, total obesity and visceral adiposity are not associated with a greater deposit of AGE. The elevation of AGE in obesity is related to the presence of cardiometabolic risk.

**Keywords:** adipose tissue; advanced glycation end-products; body composition; cardiometabolic risk; cardiovascular risk factors; novel targets; obesity; skin autofluorescence

**Citation:** Sánchez, E.; Sánchez, M.; López-Cano, C.; Bermúdez-López, M.; Valdivielso, J.M.; Farràs-Sallés, C.; Pamplona, R.; Torres, G.; Mauricio, D.; Castro, E.; et al. Is There a Link between Obesity Indices and Skin Autofluorescence? A Response from the ILERVAS Project. *Nutrients* **2023**, *15*, 203. https://doi.org/10.3390/ nu15010203

Academic Editor: Javier Gómez-Ambrosi

Received: 17 December 2022 Revised: 26 December 2022 Accepted: 27 December 2022 Published: 31 December 2022

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

#### **1. Introduction**

Obesity is a multifactorial chronic disease that can shorten the quality of life and life expectancy of patients due to its high morbidity and mortality [1]. While this is clearly established, population studies have revealed that more than 30% of patients with obesity do not present associated metabolic pathology, which has given rise to the concept of "metabolically healthy obesity" [2]. However, these patients have shown a higher risk of both diabetes and cardiometabolic disease in the medium–long term. It has been hypothesized that this is a possible initial phase prior to the development of comorbidities [3–5].

To date, the trigger for the development of these comorbidities is unknown. However, among the different hypotheses, the possible role of inflammatory adaptation against tissue hypoxia produced by the expansion of white adipose tissue is becoming increasingly relevant [6–8]. This continuous hypoxia facilitates a change towards a proinflammatory profile that enhances the secretion of cytokines such as tumor necrosis factor alpha, interleukin-6 or hypoxia inducible factor type 1 with the consequent increase in acute phase indicators such as C-reactive protein and fibrinogen [6,9,10]. These factors are related to the appearance of both local and systemic insulin resistance, endothelial dysfunction and arteriosclerosis, as well as a higher rate of cardiovascular events [11,12].

Hypoxia, along with a proinflammatory pattern and oxidative stress, are common features of obesity, all of which have been associated with increased protein glycation [5]. Taken together, increased advanced glycation end-products (AGEs) have been related to the formation of atherosclerotic plaques and increased cardiovascular risk [13,14]. Lifeline cohort studies have shown that increased AGEs are independently related to BMI, age and HbA1c level [15]. Others have shown its increase in patients with visceral obesity, related to an increased prevalence of metabolic syndrome [16]. Similarly, our group, has previously published that the increase in AGE concentration in patients with severe obesity is clearly at the expense of those with metabolic syndrome, suggesting its determination as a way of identifying those patients with "metabolically diseased obesity" [17]. However, we are missing a study specifically designed to assess the impact of obesity, as measured by both BMI and body fat, on AGE levels. With this objective, and to verify if the accumulation of AGEs could help us to identify early and easily those people with a higher risk of metabolic syndrome, we have analyzed the population of the ILERVAS project. This large cohort included subjects with one or more cardiometabolic risk factors and different weight ranges.

#### **2. Materials and Methods**

#### *2.1. Study Design*

In this work, we analyze the information collected in the ILERVAS project (ClinTrials.gov Identifier: NCT03228459), a prospective study whose main goal was learning the prevalence of non-clinical atheromatous disease and occult kidney disease in a cohort with moderate cardiovascular risk [18,19]. Data were analyzed from 4254 people recruited between 2015 and 2018. Patients were recruited aged 45 to 70 years, with no previous cardiovascular event but at least one cardiometabolic risk factor (obesity, hypertension, dyslipidemia, smoking or first-degree relative with prematurity (<55 years in men, <65 in women) cardiovascular disease (myocardial infarction, stroke and peripheral arterial disease)). Those with diabetes, chronic kidney disease, active neoplasia, a life expectancy of less than 18 months and/or pregnancy were excluded.

The ILERVAS project protocol was approved by the ethics committee of the Arnau de Vilanova University Hospital (CEIC-1410) and written informed consent was acquired from all subjects. The ethical guidelines of the Declaration of Helsinki and Spanish legislation on the protection of personal data were also followed.

#### *2.2. Definition of Cardiovascular Risk Factors*

The diagnosis of dyslipidemia was obtained from patients who had an assigned code for disorders of lipoprotein metabolism and other lipidemias by means of the International Classification of Diseases (ICD-10) codes, namely E78.0–78.9 (pure hypercholesterolemia, pure hyperglyceridemia, mixed hyperlipidemia, hyperchylomicronemia, other hyperlipidemia, unspecified hyperlipidemia, lipoprotein deficiency, other disorders of lipoprotein metabolism and unspecified disorders of lipoprotein metabolism). A diagnosis of hypertension was obtained from patients coded for hypertensive diseases using ICD-10 codes, i.e., I10–I13 (essential hypertension, hypertensive heart disease, hypertensive renal disease and hypertensive heart and renal disease) and I15 (secondary hypertension).

Prediabetes was defined as a glycated hemoglobin (HbA1c) level between 39 to 47 mmol/mol (5.7 to 6.4%), and normal glucose metabolism as HbA1c <39 mmol/mol (<5.7%), agreeing with the American Diabetes Association guidelines. Smoking habits (never, former or current smoker) were also considered. Smokers who quit smoking a year or more before the visit were considered ex-smokers. As patients with diabetes were excluded from the ILERVAS project, this diagnosis was not considered a cardiovascular risk factor in our study.

The antihypertensive and lipid-lowering treatments that were prescribed in the IL-ERVAS population have been taken from the prescription and billing databases provided by CatSalut (Catalan Health Service), which were incorporated annually into the SIDIAP database. Antihypertensive medications include angiotensin-converting enzyme inhibitors, diuretics, type II aldosterone receptor antagonists, beta-blockers, calcium channel blockers and other antihypertensives. Lipid-lowering drugs included statins, fibrates, ezetimibe and omega-3 fatty acids.

#### *2.3. Anthropometric Measures*

Both weight and height were analyzed almost without clothing and without shoes with a precision of 0.5 kg and 1.0 cm, respectively [20]. Waist circumference was measured between the iliac crest and the lower rib in the horizontal plane with the subject standing and with a non-elastic tape to a precision of 0.1 cm [21]. To decrease interobserver and device variability, all anthropometric measures were performed by trained nurses under standardized conditions. The relative technical error of intra-rater measurement was less than 1% for height, weight and waist and circumferences.

BMI was obtained by weight (kg) divided by the square of body height (m), and obesity was classified according to clinical guidelines as BMI ≥30 kg/m2. The percentage of total body fat was estimated using the Body Adiposity Estimator of the Clínica Universidad de Navarra (CUN-BAE) using the formula: −44.988 + (0.503 × age) + (10.689 × sex) + (3.172 × BMI) − (0.026 × BMI2) + (0.181 × BMI) × sex) − (0.02 × BMI × age) − (0.005 × BMI2 × sex) + (0.00021 × BMI2 × age), where sex is 1 for women and 0 for men and age is in years [22].

For the estimation of central adiposity, in addition to waist circumference, the body roundness index was included. This index, suggested by Thomas et al., is based on a geometric model defined to quantify body circularity. Those with abdominal fat look like a perfect circle, compared to those with more linear figures. It was calculated as: WC (m)/(BMI2/3 × height (m))1/2 [23]. In addition, we evaluated the Hume index for the amount of lean mass based on the analysis of the body composition of the antipyrine dilution space through the formula: (0.29569 × weight) + (0.41813 × height) − 43.2933 [24].

#### *2.4. Skin Autofluorescence*

SAF was assessed using the AGE Reader™ device (DiagnOptics Technologies, Groningen, The Netherlands), a computerized non-invasive tool that quantifies AGE deposits in the forearm via the ultraviolet spectrum [25]. A device calibrated according to the manufacturer's recommendations was used. Three analyses were carried out in areas free of tattoos, cosmetics or with a concentration of freckles or superficial vessels, and their

mean value (arbitrary units: AU) was taken. Measurements made on the same day showed an overall Altman error rate of 5.03%, and intra-individual seasonal deviation showed an Altman error rate of 5.87% [25]. Since AGEs accumulate progressively with aging, there is a normal sum of AGEs at each age. When this number is higher than expected, the software classifies the patient as a "high AGE" individual. Therefore, participants in the ILERVAS project were classified as a group with "normal" and "high" SAF levels.

#### *2.5. Statistical Methods*

The Shapiro–Wilk test was used to estimate the normal distribution of the sample. Quantitative baseline characteristics were analyzed using the Mann–Whitney U test or Kruskal–Wallis test, and categorical characteristics using Pearson's chi-squared test. Spearman's correlation was used to assess the relationship between AGE levels and anthropometric data. Data are expressed as median and interquartile range or n (percentage). Patients were differentiated based on their elevated and normal SAF results. In addition, patients were also categorized according to the number of cardiovascular risk factors.

The evaluation of the diagnostic performance of the anthropometric formulas was carried out by analyzing the area under the receiver operating characteristic (ROC) curves and the Youden J statistic. The results of the area under the ROC curve were interpreted following the guidelines stipulated by the scientific community: excellent, between 0.9 and 1.0; good, between 0.8 and 0.9; fair, between 0.7 and 0.8; poor, between 0.6 and 0.7; and not useful, between 0.5 and 0.6. SSPS software (IBM SPSS Statistics for Windows, version 20.0., Armonk, NY, USA) was used for statistical analysis. Statistical significance was determined with a *p* value < 0.05.

#### **3. Results**

The main clinical and metabolic data according to the presence of SAF levels are shown in Table 1. The ILERVAS cohort consisted of 1115 (26.2%) individuals with elevated SAF values. This group of individuals were mainly smokers with a characteristic cardiovascular risk profile better than participants with normal SAF values. This high SAF group also received significant undertreatment with antihypertensive and lipid-lowering medications. However, no differences in the prevalence of obesity, according to BMI, were observed between groups (30.0 vs. 27.9%, *p* = 0.189). Similarly, no differences in total body fat percentage or estimated visceral adiposity and lean body mass were detected between the groups (Table 2).


**Table 1.** Central clinical and metabolic data in the ILERVAS cohort according to skin autofluorescence values.

Data are expressed as a median (interquartile range) or n (percentage). Antihypertensive drugs include angiotensin-converting enzyme (ACE) inhibitors, diuretics, angiotensin-II receptor antagonists (ARA II), betablockers, calcium antagonists and other antihypertensives. Lipid-lowering treatments involve statins, fibrates, ezetimibe and omega-3 fatty acids.


**Table 2.** Data of the anthropometric indices in all individuals according to skin autofluorescence values.

Data are expressed as a median (interquartile range). BMI: body mass index; CUN-BAE: Clínica Universidad de Navarra-Body Adiposity Estimator.

Regarding the bivariate analysis, the SAF levels showed a very slight but positive correlation with the percentage of total body fat (estimated by the CUN-BAE formula) and abdominal adiposity (estimated by body roundness index) (Table 3). These correlations disappeared when anthropometric formulas such as BMI and waist circumference were used. In addition, a negative correlation with lean body mass was also observed. In the same way, the measures related to obesity and body composition had no power to identify the patients with higher levels of SAF, being in all cases areas under the ROC curve <0.52 (Figure 1).

**Table 3.** Bivariate correlations of SAF with anthropometric formulas in the ILERVAS population.


**Figure 1.** Receiver operating characteristic (ROC) curve analysis in the ILERVAS population to assess the diagnostic accuracy of obesity indices to identify patients with higher AGEs from those with normal AGEs.

We also analyzed SAF levels according to a number of cardiometabolic risk factors (dyslipidemia, hypertension, history of smoking and prediabetes) in both the total study population and according the degree of obesity. A progressive increase in SAF levels was observed in parallel with cardiovascular risk factors in the entire population (Figure 2). There were also significant differences in terms of skin autofluorescence values according to the number of cardiometabolic risk factors in subjects with normal weight, overweight and obesity (Figure 3).

**Figure 2.** Results of the skin autofluorescence values in the entire population of the ILERVAS project according to the number of cardiovascular risk factors (a history of smoking habits, hypertension, dyslipidemia and prediabetes).

Finally, the multivariable logistic regression model in patients with obesity according to their BMI showed that male sex, degree of obesity and the presence of three or more cardiovascular risk factors (prediabetes, smoking, hypertension and dyslipidemia) were independently associated with AGE levels (Table 4).


**Table 4.** The multivariable logistic regression model for high AGEs in subjects with obesity.

**Figure 3.** Results of the skin autofluorescence values in subjects with normal weight (**A**), overweight (**B**) and obesity (**C**) according to the number of cardiovascular risk factors (history of smoking habits hypertension, dyslipidemia and prediabetes).

#### **4. Discussion**

In our middle-aged Caucasian population cohort, no significant increase in AGEs (measured as skin autofluorescence) was observed with respect to overall obesity or abdominal obesity. However, the subcutaneous deposition of AGE seems to be positively related to the prevalence of cardiometabolic risk factors, both in patients with and without obesity. Until now, when the relationship between obesity and AGE deposition has been evaluated, controversial data have been shown. For example, in a study of the child population, Lentferink et al. found a correlation of AGEs with the highest standard deviation of BMI, which disappeared when adjusting for skin type [26]. Likewise, in their study, Gogas et al. observed a positive correlation of AGEs with BMI, being higher in those with type 2 diabetes [27].

Visceral adipose tissue has been shown to have a greater inflammatory capacity, so it would be expected that this confers greater oxidative conditions favoring the formation of AGEs [28]. Despite this, Den Engelsen et al. did not observe significant differences in AGEs in those with or without central obesity measured by waist circumference. However, they demonstrated a progression: from those with healthy normal weight (1.63 ± 0.37 AU), increasing in those with abdominal obesity (1.74 ± 0.44 AU) and being even higher in subjects with abdominal obesity and comorbidities (1.87 ± 0.43 AU; *p* < 0.001) [16]. In the same study, after a medium follow-up period of 3 years after bariatric surgery, the SAF values did not change, although there was a marked reduction in weight and remission of comorbidities.

Elevated AGE levels have been linked to increased cardiometabolic risk, coronary artery disease and cardiovascular mortality [14,29]. Similarly, increased subcutaneous AGE content has previously been associated with increased atheromatous plaque burden in the ILERVAS project [13]. In the present study, we found differences when we assessed patients according to their cardiovascular risk, with SAF values that progressively increased according to the accumulation of cardiometabolic risk factors. Our results are in line with those of Koyama et al. who found a significant relationship between AGE receptors and metabolic syndrome, blood pressure, hypertriglyceridemia, and subclinical atheromatosis in both patients with and without diabetes [30]. Reinforcing the role of metabolic control in AGE deposition, plasma AGEs were higher in patients with type 2 diabetes and atherosclerotic disease than in patient with atherosclerotic disease without type 2 diabetes, especially in those with higher HbA1c levels in recent years (r = 0.46, *p* < 0.001) [31].

Other cardiovascular risk factors not evaluated in our study, such as chronic kidney disease or adherence to the Mediterranean diet, also cause an increase in AGE concentration [32]. In fact, studies based on dietary surveys have associated a low intake of exogenous AGEs with lower insulin resistance, TNF alpha levels, peripheral cell mononuclear cells and leptin concentration, as well as higher adiponectin, which ultimately means less proinflammatory activity [33]. Thus, adherence to the Mediterranean diet, an eating pattern associated with lower proinflammatory state, has been independently associated with AGEs, especially in those with a high consumption of vegetables, fruits and low sugar [34]. In our investigation it is also interesting to note that for the first time the negative but statistically significant link between the levels of SAF and lean body mass is shown.

Our research has some limitations. First, we do not use a precise measure of body composition to correlate with AGEs. However, anthropometric formulas have been validated with other gold standard tests such as dual-energy X-ray absorptiometry or magnetic resonance imaging. Second, we used an indirect test based on skin fluorescence to measure AGEs instead of a direct plasma test, but there is an extensive literature demonstrating the accuracy of this test compared to skin biopsy or plasma measurements. Third, an intrinsic characteristic of the ILERVAS study population is that participants have one or more cardiometabolic risk factors, so care must be taken when generalizing our results to the general population.

#### **5. Conclusions**

In conclusion, total obesity and visceral adiposity are not associated with a higher AGE deposit. The elevated levels of AGEs detected in subjects with obesity seem more related to the presence of cardiometabolic risk factors than to the percentage of body fat. With all this evidence, the measurement of SAF is a non-invasive test that can be helpful to identify those patients with unhealthy obesity, which opens the door to a new management of obesity in clinical practice.

**Author Contributions:** Conceptualization, E.S., M.S. and A.L.; Data curation, E.S., M.S. and M.B.-L.; Formal analysis, E.S., M.S., C.L.-C. and A.L.; Funding acquisition, J.M.V., E.F. and M.B.-L.; Investigation, E.C. and J.M.V.; Methodology, C.F.-S., G.T. and M.B.-L., Software, C.F.-S., E.C.; Project administration, M.B.-L., J.M.V. and A.L.; Resources, M.B.-L., E.F. and A.L.; Supervision, R.P., and D.M.; Visualization, M.B.-L. and J.M.V.; Writing—original draft, E.S., M.S. and C.L.-C.; Writing—review & editing, A.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This investigation was sustained by grants from the Generalitat de Catalunya and Diputació de Lleida (SLT0021600250 and 2017SGR696). CIBER de Diabetes y Enfermedades Metabólicas Asociadas and CIBER de Enfermedades Respiratorias are initiatives of the Instituto de Salud Carlos III.

**Institutional Review Board Statement:** The research was conducted agreeing to the guidelines of the Declaration of Helsinki and accepted by Arnau de Vilanova University Hospital Ethics Committee (CEIC-1410, 19 December 2014).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The evidence presented in this investigation are accessible on request from the corresponding author. The data are not publicly presented due to the signed consent agreements around data sharing, which only allow access to the researchers of the ILERVAS project following the project purposes. Requestors wishing to access the data used in this work can make a demand to A.L. and M.B.-L. The request will be subjected to authorization and formal agreements regarding confidentiality and secure data storage being signed the data would be the provided.

**Acknowledgments:** The authors would like to thank to ILERVAS Project collaborators: Marta Hernández, Ferran Rius, Ferrán Barbé, Pere Godoy, Manuel Portero-Otin, Mariona Jové, Marta Ortega, Eva Miquel, Montserrat Martínez-Alonso, Jordi de Batlle, Silvia Barril, Manuel Sánchez-de-la-Torre, Josep Franch-Nadal and Esmeralda Castellblanco. We would like also to thank to more than a few nurses and staff of the project, Fundació Renal Jaume Arnó, and the Primary Care teams of the province of Lleida for enrolling individuals and their energies in the correct progress of the study. The funders had no role in work design, the collection, analysis and interpretation of data, report writing, or the decision to submit the article for publication.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


Ecology. *Nutrients* **2019**, *11*, 215. [CrossRef] [PubMed]

33. Baye, E.; Kiriakova, V.; Uribarri, J.; Moran, L.J.; de Courten, B. Consumption of diets with low advanced glycation end products improves cardiometabolic parameters: Meta-analysis of randomised controlled trials. *Sci. Rep.* **2017**, *7*, 2266. [CrossRef] [PubMed] 34. Snelson, M.; Coughlan, M.T. Dietary Advanced Glycation End Products: Digestion, Metabolism and Modulation of Gut Microbial

**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
