Dietary advanced glycation end-product restriction for the attenuation of insulin resistance, oxidative stress and endothelial dysfunction: a systematic review


The benefits of advanced glycation end-product (AGE)-restricted diets in humans are unclear. This review aimed to determine the effect of dietary AGE restriction on the inflammatory profiles of healthy adults and adults with diabetes or renal failure. Eight computer databases were searched for controlled feeding trials published in English between January 1997 and December 2012. Human trials were included if at least one group received an AGE-restricted dietary intervention. A total of 12 trials reporting on 289 participants were included in the review. Five trials (42%) were of high methodological quality. Meta-analysis of two long-term (16 week) trials provided evidence favoring an AGE-restricted diet for the reduction of 8-isoprostanes (standardized mean difference 0.9; 95% confidence interval (CI): 0.3–1.5) and tumor necrosis factor-α (1.3; 95% CI: 0.6–1.9) in healthy adults. Intermediate-term dietary AGE restriction in adults with chronic renal failure reduced serum VCAM-1 (0.9; 95% CI: 0.1–1.7). Individual trials provided some evidence that long-term dietary AGE restriction reduces HOMA-IR (1.4; 95% CI: 0.3–2.6) and AGE-modified low-density lipoprotein (2.7; 95% CI: 1.6–3.9) in adults with type 2 diabetes. Generalisability is limited, as 75% of studies were of less than 6 weeks duration and more than half were of low methodological quality. Evidence quality ranged from low to very low, limiting the conclusions that can be drawn from this review. There is currently insufficient evidence to recommend dietary AGE restriction for the alleviation of the proinflammatory milieu in healthy individuals and patients with diabetes or renal failure. Additional long-term high-quality RCTs with larger sample sizes measuring patient-important outcomes are required to strengthen the evidence supporting the effects of AGE-restricted diets.


Advanced glycation end products (AGEs) are formed endogenously when the carbonyl groups of reducing sugars nonenzymatically react with the free amino groups on proteins. AGEs are generated in vivo as a normal consequence of metabolism, but their formation is accelerated under conditions of hyperglycemia, hyperlipidemia and increased oxidative stress.

Although glucose is relatively slow in reacting with proteins, highly reactive dicarbonyl compounds (generated as a result of glucose auto-oxidation, lipid peroxidation and the interruption of glycolysis by reactive oxygen species) are capable of rapid AGE formation. Dicarbonyls such as glyoxal, methylglyoxal and 3-deoxyglucosone interact with intracellular proteins to form AGEs, and can also diffuse out of the cell and react with extracellular proteins.

Excessive AGE accumulation results in significant cellular dysfunction by inhibiting communication between cells, altering protein structure and interfering with lipid accumulation within the arterial wall.1 Interaction of AGEs with the receptor for AGEs (RAGE) activates nuclear factor κB, triggering oxidative stress, thrombogenesis, vascular inflammation and pathological angiogenesis,2 thereby contributing to many of the long-term complications of diabetes. More recently, AGEs have been implicated in the pathogenesis of type 2 diabetes by contributing to the development of insulin resistance and low-grade inflammation known to precede the condition.3, 4

Apart from endogenous AGE formation, AGEs and their precursors are also absorbed by the body from exogenous sources such as cigarette smoke and through consumption of highly heated processed foods. Browning of food during cooking is used to enhance the quality, flavour, color and aroma of the diet. This process (known as the Maillard reaction) generates large quantities of AGEs.5 Factors that enhance AGE formation in foods include high lipid and protein content, low water content during cooking, elevated pH and the application of high temperature over a short time period. More AGEs are generated in foods exposed to dry heat (grilling, frying, roasting, baking and barbecuing) than foods cooked at lower temperatures for longer time periods in the presence of higher water content (boiling, steaming, poaching, stewing or slow cooking).6

Kinetic studies have demonstrated that approximately 10–30% of dietary AGEs consumed are intestinally absorbed,7 with only one-third of ingested AGEs excreted in urine and feces. Plasma AGE concentration appears to be directly influenced by dietary AGE intake and the body’s capacity for AGE elimination.8 Individuals with renal insufficiency demonstrate reduced urinary excretion of dietary AGEs, and plasma AGE levels inversely correlate with renal function.9

Low-AGE diets in animal studies have been shown to reverse insulin resistance and chronic inflammation, inhibit the progression of atherosclerosis and prevent experimental diabetic nephropathy and neuropathy,10 but whether these results can be translated to humans is uncertain. Cross-sectional and case–control studies involving humans with impaired renal function or diabetes have demonstrated associations between elevated AGE intakes and serum biomarkers of oxidative stress, endothelial dysfunction, inflammation, hyperlipidemia and hyperglycemia.11, 12 AGEs have also recently been implicated in the dysfunction and death of pancreatic beta cells,13 leading to the hypothesis that excessive AGE formation and oxidative stress possibly have a role in the development of type 1 and type 2 diabetes.14, 15 Low-AGE diets have been suggested as a possible future therapeutic option for healthy individuals at risk for the development of type 1 or type 2 diabetes.16

Through reduced consumption of highly processed heat-treated foods, dietary AGE restriction may represent a relatively simple, noninvasive therapy for the effective treatment of many of the metabolic disturbances attributed to excessive AGE levels. This systematic review sought to determine whether there is sufficient evidence to recommend therapeutic AGE-restricted diets in healthy or overweight individuals, people with diabetes or those with renal impairment for the prevention or attenuation of insulin resistance, the improvement of endothelial function and the reduction of biomarkers of inflammation and oxidative stress.

Materials and methods

Search strategy

A computer database search was undertaken for the time period between 1 January 1 1997 and 1 December 2012, using Medline, CINAHL, EMBASE, Current Contents, PubMed, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews and AMED. Databases were not searched before 1997, because the potentially deleterious effects of dietary AGE consumption was first postulated in 1997.17 Citation tracking was performed using the ISI Web of Science for all trials identified, and the reference lists of all identified trials were hand-searched for relevant studies. The following search terms were used: (1) (diet$ OR food) and (advanced glyc$ OR glycation OR Maillard OR thermal), (2) limit 1 to year=‘1997–2012’, (3) limit 2 to humans.

Criteria for selecting trials in this review

All full reports of controlled feeding trials were eligible for inclusion if they were published in English between 1 January 1997 and 1 December 2012. Trials were included if they involved human participants aged 18 years and at least one group of participants received an AGE-restricted dietary intervention. For the purposes of this review, we defined a low-AGE dietary intervention as one that contained 30–50% of the measured AGEs or Maillard reaction products (MRPs) present in the standard or high-AGE comparison diet. Trials involving dietary restriction of AGE precursors only (such as Amadori products) were not included.

The outcomes of interest in this review were serum markers of the following: (1) insulin resistance (HOMA-IR), (2) inflammation (tumor necrosis factor-α (TNF-α)), (3) oxidative stress (8-isoprostane), (4) endothelial dysfunction (VCAM-1) and (5) increased cardiovascular disease risk (AGE-modified low-density lipoprotein (LDL)). On the basis of the duration of low-AGE dietary interventions used in the included trials, the length of follow-up of outcomes was categorized as short term (one meal to 6 days after randomization), intermediate term (1–4 weeks after randomization) or long term (more than four weeks after randomization).

Assessment of methodological quality

The two reviewers independently assessed the methodological quality of included trials using the Heyland Methodological Quality Score18 (Supplementary Table S1). This checklist rates primary research based on the use of allocation concealment during randomization, intention-to-treat analysis, double-blinding, patient selection with minimal risk of bias, comparability of intervention and control groups at baseline, 100% participant follow-up, clearly described treatment protocol and well-defined outcome measurements. Trials scoring 8 out of a possible 14 points are considered to be of high methodological quality. Disagreements between reviewers in assigning methodological quality scores were resolved by discussion until consensus was achieved.

Data extraction and analysis

Trial information regarding the type and number of participants, interventions used and significant findings was extracted from each study by the first author and entered into a standardized computer spreadsheet. As all extracted data were continuous, treatment effects and 95% confidence intervals (CIs) were calculated using the Hedges (adjusted-g) standardized mean difference (SMD).19 The ‘adjusted’ statistic was used because it includes an adjustment for bias from small sample sizes. The SMD enables comparison of effect sizes between trials that use different outcome measures.20 SMDs were calculated from group mean results and s.d’s) collected at the time of follow-up. When mean values were not available, trial authors were contacted to provide the appropriate data. When standard errors were reported, these were converted to s.d’s as per Cochrane Collaboration Guidelines.21 SMDs were standardized so that positive values indicated effects favoring the AGE-restricted dietary intervention, and negative values were used to indicate effects favoring the standard diet. SMD values of 0.2, 0.5 and 0.8 were considered to represent small, moderate and large effect sizes, respectively.22

Data synthesis

Meta-analysis of pooled data was implemented in cases in which at least two trials contained similar participants (health status), intervention (low-AGE diet), comparison intervention (standard-AGE diet), outcome measures and length of follow-up. Trials with similar characteristics were assessed for statistical heterogeneity, which was indicated by a P<0.1 on the χ2 test and an I2 statistic greater than 20%.23 Clinically and statistically homogeneous trials (I2<20%) underwent a fixed-effects model meta-analysis using RevMan 5.1.24

Where meta-analysis was considered not possible because of clinical or statistical heterogeneity, effect sizes and 95% CIs were reported for outcomes within individual trials, and a narrative analysis was performed using the GRADE (Grades of Recommendation, Assessment, Development & Evaluation) approach for collating evidence in systematic reviews25 (Supplementary Table S2). The GRADE criteria consider randomized controlled trials as high-quality evidence, which can be downgraded to moderate-, low- or very low-quality evidence in the event of limitations to methodological quality (defined in this review as a Heyland Methodological Quality Score <8), inconsistency of results between trials, imprecision of results due to small sample sizes and/or wide CIs, indirectness of results due to the measurement of secondary end points or a high probability of reporting bias.


Description of selected trials

A total of 3855 citations were originally identified at the time of the initial database search, and progressed through each stage of the selection process according to the predefined inclusion criteria (Figure 1). Sixteen articles reporting on 12 controlled trials including 289 participants were ultimately included in the review.17, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 The characteristics of the included trials are outlined in Table 1.

Figure 1

Flowchart showing progression of trials through the selection process.

Table 1 Characteristics of included studies

Of the 12 trials included in this review, 4 trials included only healthy participants,26, 28, 29, 35 1 trial included only overweight or obese participants,30 3 trials included only participants with type 1 or type 2 diabetes27, 31, 39 and 1 trial included only participants with nondiabetic renal failure receiving peritoneal dialysis.37 The remaining three trials included combinations of healthy participants, patients with diabetes and/or patients with chronic renal disease.17, 38, 40

Four trials reported data on short-term low-AGE dietary interventions,17, 29, 31, 35 five trials presented intermediate-term follow-up data26, 28, 30, 37, 39 and three trials contained long-term follow-up data.27, 38, 40

A variety of post-intervention outcome measures were reported by the trials in this review, including differences in serum AGE concentration,17, 26, 27, 30, 34, 35, 36, 38, 39, 40 serum markers of insulin resistance (homeostasis model of assessment, fasting insulin, adiponectin),26, 32, 38 indicators of inflammatory processes (C-reactive protein, TNF-α, macrophage migration inhibitory factor)30, 37, 38, 39, 40 and oxidative stress (thiobarbituric acid reactive substances, 8-isoprostanes, leptin, nuclear factor κB).30, 31, 35, 38, 40 Some studies measured surrogate biochemical markers of endothelial dysfunction (E-selectin, vascular cell adhesion molecule-1, intracellular adhesion molecule-1, monocyte chemoattractant protein-1),30, 31, 37, 39, 40 cardiovascular disease risk factors (AGE-modified low-density lipoprotein, flow-mediated dilatation, plasminogen activator inhibitor-1),27, 28, 31, 36, 39 anti-inflammatory molecules (AGE receptor 1, vitamin C, vitamin E, ubiquinol)26, 38, 40 and plasma lipid levels.26

Methodological quality of trials

Methodological quality ratings for each trial according to the Heyland Methodological Quality Score are presented in Table 1. Five trials (42%) were considered to be of high quality.26, 27, 30, 31, 40 Common methodological limitations included failure to blind researchers,17, 26, 27, 28, 29, 35, 37, 38, 39, 40 inadequate randomization 17, 28, 29, 35, 39 and failure to report an intention-to-treat analysis.17, 26, 27, 28, 29, 30, 31, 35, 36, 38, 39 All of the trials failed to provide sample size calculations or explain methods of allocation concealment, and few commented on the validity and reliability of their chosen outcome measures. Most studies failed to clearly distinguish primary research outcomes from secondary research outcomes.17, 26, 27, 30, 31, 36, 38, 39, 40 Two out of 12 studies stated a clear number and reason for withdrawals,26, 36 with the remaining ten trial reports not mentioning withdrawals. Most of the trials were of limited length, with 75% of studies less than 6 weeks duration. Small sample sizes in the majority of trials resulted in outcomes with wide CIs, introducing possible uncertainties regarding the precision of the findings.

Change in serum carboxymethyl-lysine

A summary of results of this review are presented in Table 2. Ten trials measured serum carboxymethyl-lysine (CML) concentrations after individuals had received low-AGE and standard-AGE diets (Figure 2). Two short-term trials of variable quality33, 35 were unable to detect any difference in serum CML levels after a single low-AGE or standard-AGE meal, in either healthy subjects (n=9) or patients with type 2 diabetes (n=20). These results conflicted with another short-term trial,17 which measured serum CML in a sample of patients with type 1 and type 2 diabetes (n=15). Pooled serum samples collected for 48 h after consumption of a low-AGE meal contained substantially lower CML levels than serum collected after a high-AGE meal (SMD 1.8; 95% CI: 0.5–3.1).

Table 2 Summary of results according to the GRADE criteria
Figure 2

Effects of low-AGE diets on circulating CML concentration. Forest plot of standardised mean differences (95% CI) for individual and homogeneous pooled trials.

The effects of intermediate-term dietary AGE restriction on serum CML levels are mixed. In one high-quality trial,30 circulating CML levels were significantly increased after overweight but otherwise healthy volunteers (n=11) had consumed an AGE-restricted diet for 2 weeks. However, another trial of lower quality39 supported the use of a 2-week low-AGE diet in patients with diabetes to significantly reduce serum CML concentration (SMD 1.1; 95% CI: 0.2–2.0). A high-quality investigation of serum CML levels after a 4-week low-AGE intervention in 64 healthy subjects26 demonstrated a small but nonsignificant reduction in serum CML (SMD 0.3; 95% CI: −0.05 to 0.7). Two trials measured circulating CML concentrations in adults with renal failure after a 4-week consumption of either low- or standard-AGE diets.38, 40 These trials were determined to be clinically and statistically homogeneous (I2=0%), and a meta-analysis was performed using a fixed-effect model (n=27). The pooled SMD for CML was 0.5 (95% CI: −0.2 to 1.3), indicating a nonsignificant reduction in serum CML concentration after the low-AGE intervention.

Two trials investigated the effects of long-term (16 weeks) low-AGE diets versus standard-AGE diets on the serum CML levels of healthy volunteers.38, 40 These trials were clinically and statistically homogeneous (I2=0%), and thus were subjected to meta-analysis using a fixed-effect model (n=48). The pooled SMD for serum CML was 1.2 (95% CI:0.5–1.8), indicating a statistically significant effect supporting long-term dietary AGE restriction for reducing circulating CML concentrations in healthy adults. In addition, meta-analysis was performed on two homogeneous trials (I2=0%) involving long-term (6–16 weeks) low or standard-AGE dietary interventions in adults (n=42) with type 2 diabetes.27, 38 The pooled SMD for serum CML was 2.0 (95% CI: 1.2–2.8), providing low-quality evidence that long-term low-AGE diets reduce circulating CML concentrations in people with type 2 diabetes.

Healthy subjects

Two trials of variable quality compared the effects of low-AGE and high-AGE diets on serum 8-isoprostanes and TNFα in healthy adult volunteers after 16 weeks.38, 40 These trials were determined to be clinically and statistically homogeneous (I2=16% and I2=0%, respectively), and meta-analyses were performed using a fixed-effect model (n=48). The pooled SMD for serum 8-isoprostanes was 0.9 (95% CI:0.3–1.5), indicating a statistically significant effect favoring long-term low-AGE dietary intake over standard AGE intake. Similarly, the pooled SMD for TNFα was 1.3 (95% CI: 0.6–1.9), indicating large reductions in this inflammatory cytokine after the low-AGE intervention (Figure 3). One high-quality trial30 involving an intermediate-term low-AGE diet in overweight/obese individuals who were otherwise healthy (2 weeks, n=11) found a small reduction in urinary 8-isoprostane excretion after the AGE-restricted diet when compared with the standard-AGE diet (SMD 0.5; 95% CI: −0.4 to 1.3), but the difference was not statistically significant.

Figure 3

Effects of low-AGE diets on circulating biomarkers. Forest plot of standardised mean differences (95% CI) for individual and homogeneous pooled trials.

Low-quality evidence in one high-quality trial40 supported the use of a long-term (16 week) AGE-restricted diet in healthy volunteers (n=30) to significantly reduce the level of VCAM-1 (a metabolic marker of endothelial dysfunction), with an SMD of 0.9 (95% CI: 0.2–1.7).

Conflicting evidence was found on the intermediate-term (4 weeks, n=64) and long-term (16 weeks, n=18) effects of low-AGE diets on insulin resistance (measured by HOMA) in healthy individuals. Findings were inconsistent among two trials, with one high-quality trial finding increased insulin sensitivity after 4 weeks, with an SMD of 3.8 (95% CI: 3.3–4.3),26 and the other lower-quality trial finding no statistically significant effect on insulin sensitivity after 16 weeks, with an SMD of 0.7 (95% CI: −1.6 to 0.3).38

Patients with diabetes

All included trials investigating the effect of dietary AGE restriction on patients with type 1 and/or type 2 diabetes were clinically heterogeneous, and thus meta-analyses could not be performed (Figure 3). The results of one low-quality trial provided very low-quality evidence that a long-term (16 weeks, n=18) low-AGE diet reduced insulin resistance (measured by HOMA) in individuals with type 2 diabetes, with an SMD of 1.4 (95% CI: 0.3–2.6).38

In one high-quality trial,31 short-term (2 h post-prandial, n=20) comparison of low-AGE and standard-AGE meals in patients with type 2 diabetes failed to show a statistically significant difference in VCAM-1 levels, with an SMD of 0.4 (95% CI: −0.2 to 1.0) (data not shown in Forest Plot). However, one low-quality study39 demonstrated a large reduction in VCAM-1 after an intermediate-term (2 weeks, n=11) low-AGE diet in patients with type 1 and type 2 diabetes, with an SMD of 1.0 (95% CI: 0.1–1.9).

Two lower-quality trials38, 39 assessed the intermediate- and long-term effects of dietary AGE restriction on TNFα in patients with diabetes. The intermediate-term intervention (2 weeks, n=11) did not show a statistically significant result (SMD 0.3; 95% CI: −0.6 to 1.1), but the long-term intervention (16 weeks, n=18) significantly reduced serum TNFα levels (SMD 1.7; 95% CI: 0.6–2.9). The same long-term intervention significantly reduced serum 8-isoprostane levels (SMD 1.4; 95% CI: 0.3–2.5).

Similarly, AGE-modified LDL (a serum marker associated with cardiovascular disease risk) was moderately reduced by an intermediate-term AGE-restricted diet (2 weeks, n=11) in people with diabetes, but the result was not statistically significant, with an SMD of 0.8 (95% CI: −0.04 to 1.7).39 After long-term dietary AGE restriction (6 weeks, n=24), however, the reduction in AGE-modified LDL was highly significant in a high-quality trial, with an SMD of 2.7 (95% CI: 1.6–3.9).27

Patients with renal failure

Two trials of variable quality investigated the effects of intermediate-term (4 weeks n=27) low-AGE diets versus standard-AGE diets on VCAM-1 and TNFα in patients with renal failure.37, 40 These trials were clinically and statistically homogeneous (I2=0% for both variables), and thus were subjected to meta-analyses using a fixed-effect model. The pooled SMD for serum VCAM-1 was 0.9 (95% CI: 0.1–1.7), indicating a statistically significant effect favoring an intermediate-term low-AGE dietary intake on endothelial function (Figure 3). However, the pooled SMD for TNFα after 4 weeks of a low-AGE diet was not statistically significant, 0.5 (95% CI: −0.3 to 1.3).

A 4-week low-AGE dietary intervention40 in nine patients with renal failure was unable to demonstrate a statistically significant reduction in serum 8-isoprostanes when compared with those receiving the standard-AGE diet: SMD, 0.2 (95% CI: −1.1 to 1.6). Another intermediate-term (4 weeks, n=18) trial provided very low-level evidence that a 4-week low-AGE diet in people with renal failure was long enough to significantly reduce AGE-modified LDL levels, with an SMD of 1.5 (95% CI: 0.4–2.6).37

Long-term low-AGE dietary intervention trials have not been performed in patients with renal failure.

Adverse effects of low-age diets

Some AGEs (melanoidins, aminoreductones and heterocyclic compounds) demonstrate antioxidant activity, and thus their reduction in the diet may be deleterious in the longer term. One low-quality trial reported a negative consequence of an intermediate-term low-AGE diet in healthy volunteers (1 week, n=8).28 This trial found that consumption of a diet rich in MRPs significantly increased the resistance of plasma LDL to oxidation in vitro when compared with consumption of a diet low in MRPs. Results demonstrated a negative effect of low-AGE diets on the oxidative resistance of LDL, with an SMD of −4.4 (95% CI: −7.1 to −1.9). No other adverse effects of low-AGE diets were reported by the trials in this review. In contrast, one high-quality trial26 found higher concentrations of a number of serum antioxidants after a 4-week low-AGE diet compared with a standard-AGE diet in 64 healthy individuals. The SMD for post-intervention vitamin C was 2.5 (95% CI: 2.1–2.9).


Efficacy of dietary age restriction

This review provides some preliminary evidence suggesting that utilization of an AGE-restricted diet might be a successful long-term intervention to reduce the body’s total AGE concentration in both healthy individuals and people with type 2 diabetes. However, caution is required in the interpretation of these results. All of the individual and pooled trials showing a positive effect of low-AGE diets on reducing systemic CML levels were performed by the same research team, and while a proportion of their trials were of high quality according to the Heyland Methodological Quality Score, these studies need to be replicated by other research groups to strengthen the evidence. There has been no direct determination of the proportion of exogenous dietary AGEs or their precursors which contribute to serum CML levels, and what proportion is a result of endogenous glycation. Some studies have found that energy-restricted41 and antioxidant-supplemented42 (rather than AGE-restricted) diets also reduce serum CML concentrations, indicating that other factors such as body fat, circulating triglycerides and antioxidant capacity may have additional effects on serum CML levels. The rate of endogenous protein degradation and turnover is also likely to influence circulating CML concentrations. Publication bias may be a factor to consider in this particular area of research, as published cross-sectional studies exist that fail to detect any correlation between dietary AGE consumption and circulating CML levels.43, 44, 45 Other studies with similar negative findings may remain unpublished. Finally, CML is only one of many AGEs, and further study is required to determine the effects of a low-AGE diet on other circulating AGEs and what their specific functions are.

This review found no evidence that an AGE-restricted diet reduces serum CML in individuals with renal failure. However, it is well established that kidney dysfunction impairs AGE excretion, increasing the level of circulating AGEs.46

This review found low-quality evidence supporting adherence to a long-term low-AGE diet for the reduction of 8-isoprostanes and TNFα (biomarkers of oxidative stress and inflammation, respectively) in healthy individuals. Very low-quality evidence supports long-term dietary AGE restriction for the reduction of 8-isoprostanes in people with type 2 diabetes. These findings are consistent with large cross-sectional studies conducted in healthy individuals10 and patients with type 2 diabetes.12

Contradictory evidence was found in this review regarding the effect of a low-AGE diet on markers of insulin resistance (HOMA) in healthy individuals. This may be because people with a healthy body weight are generally insulin sensitive to begin with, and interventions to reduce insulin resistance are therefore of limited benefit. The evidence favoring a long-term low-AGE diet for the attenuation of insulin resistance in people with type 2 diabetes was of very low quality.

Low-quality evidence in this review supported an intermediate-term AGE-restricted diet for the reduction of VCAM-1 in patients with renal failure. The effects of dietary AGE restriction on VCAM-1 levels in people with diabetes were contradictory. A long-term AGE-restricted diet reduced VCAM-1 levels in healthy individuals compared with a standard AGE diet. This review also found low-quality evidence in favor of long-term dietary AGE restriction for the reduction of AGE-modified LDL in patients with type 2 diabetes. Very low-quality evidence supported an intermediate-term low-AGE diet for the reduction of AGE-modified LDL in patients with renal failure. Cross-sectional studies support an association between Western-style dietary patterns and the pathogenesis of cardiovascular disease;47, 48 however, AGEs are likely to be only one of multiple dietary components involved.

Methodological limitations of included trials

Many of the trials included in this study contained major methodological flaws, with less than half receiving high-quality scores based on the Heyland Methodological Quality criteria. Few studies described the method of randomization used, and the method of allocation concealment was not mentioned in any of the trials. Most of the trials contained very small numbers of participants, and were likely to be underpowered. Primary research outcomes were not clearly defined in most of the trials, making it difficult to determine whether sample sizes were adequate.

Eight of the 12 trials included in this review estimated the AGE content of their test diets based on a database of the AGE content of common foods initially published by Goldberg et al.,49 and later updated by Uribarri et al.50 This research team utilized a nonvalidated, enzyme-linked immunosorbent assay-based method for the measurement of CML in foodstuffs. Immunological methods of AGE measurement have been associated with some limitations, as the assays are capable of detecting additional contaminants that are unrelated to the ligand of interest.51 An additional disadvantage of this indirect, enzyme-linked immunosorbent assay method is that it only allows the AGE concentration in foods to be expressed in arbitrary units (kilounits AGE), making comparisons with other analytical techniques impossible. There is a need for all researchers in this field to use standardized, validated measurement tools for the assessment of the AGE concentration in foods. Liquid chromatography–mass spectrometry methods appear to be highly sensitive techniques for AGE quantification,52, 53 and would enable comparisons of results to be made between different laboratories.

Apart from AGEs, multiple other RAGE ligands exist and opinion is divided over the binding affinity of dietary AGEs for RAGE in vivo.54 Only highly glycated proteins appear to successfully bind to and activate RAGE in vitro,55 with the low-molecular-weight AGEs absorbed into the circulation after digestion of a high-AGE meal unlikely to interact with RAGE. Although post-absorptive dietary AGE precursors may modify endogenous proteins in order to form high-molecular-weight compounds capable of RAGE binding, this has not yet been determined. Elevations in postprandial hyperglycemia appear to be sufficient to increase the expression of nuclear factor κB56 independent of the AGE content of the meal.35

The majority of trials in this review measured serum CML as an indicator of circulating AGEs; however, the body’s total AGE concentration is currently unknown. CML is only one of many different AGEs, most of which have not yet been characterized. Although some AGEs may have a role in cellular toxicity, others may confer beneficial antioxidant effects. Individual AGEs are also likely to have different rates of absorption and excretion. A large proportion of systemic AGEs may simply be a by-product of oxidative stress and inflammation rather than a causative factor. High-quality studies are required to provide answers to the many uncertainties surrounding the function and kinetics of AGEs.

Limitations of this review

A wide variety of outcome measures were utilized by the studies included in this review, in patients with a number of different health conditions and a length of follow-up ranging from 2 h to 16 weeks. The heterogeneity among studies made it difficult to collate the findings of more than two trials for any particular outcome. Most of the evidence presented in this review was therefore derived from individual studies. More than half of the trials included in this review were determined to be of low methodological quality, with a Heyland Methodological Quality Score <8.

Because of the absence of research investigating the effects of dietary AGE restriction on patient-important primary outcomes, such as diabetes diagnosis, microvascular and macrovascular complications and quality of life, surrogate biochemical measurements (secondary end points) were evaluated in this review. Although the biochemical indices assessed in this review are considered valid metabolic markers of insulin resistance, inflammation, oxidative stress, endothelial dysfunction and cardiovascular disease risk,57 the GRADE criteria automatically downgrades the strength of evidence favoring a particular intervention from high to medium when secondary rather than primary end points are summarized.58 The strength of evidence from studies included in this review were further downgraded from medium to low quality by the GRADE criteria owing to their small sample sizes and wide CIs, indicating possible imprecision of results. Evidence from studies with Heyland Methodological Quality Scores <8 was also downgraded from low to very low quality because of methodological limitations. Future high-quality trials involving modulation of dietary AGE levels, which include the assessment of at least one primary outcome and the use of a priori sample size calculations, will enhance the strength of the current evidence base.

Seven of the 12 trials included in this review were undertaken by the same research group, possibly introducing similar methodological constraints into the majority of studies conducted on this topic.

Comparison with other reviews

No other systematic reviews have addressed the potential health benefits of AGE-restricted diets. A review of the effects of dietary factors on low-grade inflammation in overweight individuals59 concluded that despite many gaps in the research investigating the effects of dietary AGE consumption on chronic inflammation, ‘it might be prudent to advise renal failure patients to decrease their intake of ‘highly heated’ food’. However, as the effects of long-term dietary AGE restriction in people with renal failure are currently unknown, further dietary restrictions may contribute to a greater and unnecessary burden in this patient group.

The evidence summarized in this review suggests that patients with diabetes could potentially reduce their level of insulin resistance, systemic inflammation and oxidative stress and their risk of cardiovascular events by adhering to a long-term low-AGE diet. The minimization of MRP formation during food preparation and cooking is a simple, inexpensive and noninvasive intervention that could potentially alleviate significant diabetes-related morbidity. Preventing excessive heat treatment of food is also in line with current anticancer recommendations for restricting the consumption of heterocyclic amines and polycyclic aromatic hydrocarbons.60 At present, however, it cannot be ruled out that the health benefits associated with reduced consumption of highly heated food may simply be a result of an increased intake of antioxidants (which would have otherwise been destroyed during the cooking process), or avoidance of deleterious compounds other than AGEs, which are generated during thermal processing such as acrylamide or heterocyclic amines. There is a need for well-designed controlled feeding trials that evaluate effects of the consumption of dietary AGEs and their precursors in purified forms, in order to rule out contributions made by other macronutrients and micronutrients in whole foods.

Reviewers’ conclusions

Current evidence supporting the efficacy of medium- to long-term dietary AGE restriction for alleviating the proinflammatory milieu in healthy individuals and patients with diabetes or renal failure is of low quality. Additional long-term high-quality RCTs with larger sample sizes measuring patient-important primary end points are required to strengthen the evidence supporting the effects of low-AGE diets. Further studies utilizing standardized methods of AGE measurement are needed to elucidate the role of specific AGEs in cellular dysfunction, and explore possible adverse effects of low-AGE diets in the long term.

Although it is likely that future research findings will support the use of dietary AGE restriction as a therapeutic strategy to assist in the management of adults with prediabetes, diabetes and renal impairment, there is currently insufficient evidence to encourage the use of AGE-restricted diets in mainstream nutrition practice. More information is required regarding the digestion, absorption, function and elimination of dietary AGEs in their pure form, along with standardized, validated tools for the measurement of deleterious AGEs in foods.


  1. 1

    Barlovic DP, Thomas MC, Jandeleit-Dahm K . Cardiovascular disease: what’s all the AGE/RAGE about? Cardiovasc Hematol Disord Drug Targets 2010; 10: 7–15.

    CAS  Article  Google Scholar 

  2. 2

    Yamagishi S, Ueda S, Okuda S . Food-derived advanced glycation end products (AGEs): a novel therapeutic target for various disorders. Curr Pharm Des 2007; 13: 2832–2836.

    CAS  Article  Google Scholar 

  3. 3

    Bierhaus A, Humpert PM, Morcos M, Wendt T, Chavakis T, Arnold B et al. Understanding RAGE, the receptor for advanced glycation end products. J Mol Med 2005; 83: 876–886.

    CAS  Article  Google Scholar 

  4. 4

    Tahara N, Yamagishi S, Matsui T, Takeuchi M, Nitta Y, Kodama N et al. Serum levels of advanced glycation endproducts (AGEs) are independent correlates of insulin resistance in nondiabetic subjects. Cardiovasc Ther 2012; 30: 42–48.

    CAS  Article  Google Scholar 

  5. 5

    Tessier FJ, Birlouez-Aragon I . Health effects of dietary Maillard reaction products: the results of ICARE and other studies. Amino Acids 2012; 42: 1119–1131.

    CAS  Article  Google Scholar 

  6. 6

    Uribarri J, Cai W, Sandu O, Peppa M, Goldberg T, Vlassara H . Diet-derived advanced glycation end products are major contributors to the body’s AGE pool and induce inflammation in healthy subjects. Ann NY Acad Sci 2005; 1043: 461–466.

    CAS  Article  Google Scholar 

  7. 7

    Faist V, Erbersdobler HF . Metabolic transit and in vivo effects of melanoidins and precursor compounds deriving from the Maillard reaction. Ann Nutr Metab 2001; 45: 1–12.

    CAS  Article  Google Scholar 

  8. 8

    Delgado-Andrade C, Tessier FJ, Niquet-Leridon C, Seiquer I, Pilar Navarro M . Study of the urinary and faecal excretion of N (epsilon)-carboxymethyllysine in young human volunteers. Amino Acids 2012; 43: 595–602.

    CAS  Article  Google Scholar 

  9. 9

    Kankova K, Sebekova K . Genetic variability in the RAGE gene: possible implications for nutrigenetics, nutrigenomics, and understanding the susceptibility to diabetic complications. Mol Nutr Food Res 2005; 49: 700–709.

    CAS  Article  Google Scholar 

  10. 10

    Uribarri J, Cai W, Peppa M, Goodman S, Ferrucci L, Striker G et al. Circulating glycotoxins and dietary advanced glycation endproducts: two links to inflammatory response, oxidative stress, and aging. J Gerontol A Biol Sci Med Sci 2007; 62: 427–433.

    Article  Google Scholar 

  11. 11

    Uribarri J, Peppa M, Cai W, Goldberg T, Lu M, Baliga S et al. Dietary glycotoxins correlate with circulating advanced glycation end product levels in renal failure patients. Am J Kidney Dis 2003; 42: 532–538.

    CAS  Article  Google Scholar 

  12. 12

    Chao PC, Huang CN, Hsu CC, Yin MC, Guo YR . Association of dietary AGEs with circulating AGEs, glycated LDL, IL-1alpha and MCP-1 levels in type 2 diabetic patients. Eur J Nutr 2010; 49: 429–434.

    CAS  Article  Google Scholar 

  13. 13

    Ramasamy R, Yan SF, Schmidt AM . Receptor for AGE (RAGE): signaling mechanisms in the pathogenesis of diabetes and its complications. Ann N Y Acad Sci 2011; 1243: 88–102.

    CAS  Article  Google Scholar 

  14. 14

    Coughlan MT, Yap FY, Tong DC, Andrikopoulos S, Gasser A, Thallas-Bonke V et al. Advanced glycation end products are direct modulators of beta-cell function. Diabetes 2011; 60: 2523–2532.

    CAS  Article  Google Scholar 

  15. 15

    Fiory F, Lombardi A, Miele C, Giudicelli J, Beguinot F, Van Obberghen E . Methylglyoxal impairs insulin signalling and insulin action on glucose-induced insulin secretion in the pancreatic beta cell line INS-1E. Diabetologia 2011; 54: 2941–2952.

    CAS  Article  Google Scholar 

  16. 16

    Vlassara H, Striker GE . AGE restriction in diabetes mellitus: a paradigm shift. Nat Rev Endocrinol 2011; 7: 526–539.

    CAS  Article  Google Scholar 

  17. 17

    Koschinsky T, He CJ, Mitsuhashi T, Bucala R, Liu C, Buenting C et al. Orally absorbed reactive glycation products (glycotoxins): an environmental risk factor in diabetic nephropathy. Proc Natl Acad Sci USA 1997; 94: 6474–6479.

    CAS  Article  Google Scholar 

  18. 18

    Heyland DK, Novak F, Drover JW, Jain M, Su X, Suchner U . Should immunonutrition become routine in critically ill patients? A systematic review of the evidence. JAMA 2001; 286: 944–953.

    CAS  Article  Google Scholar 

  19. 19

    Hedges I, Olkin I Statistical Methods for Meta-analysis 1st edn. Academic Press: Orlando, FL, USA, 1985.

    Google Scholar 

  20. 20

    White IR, Thomas J . Standardized mean differences in individually-randomized and cluster-randomized trials, with applications to meta-analysis. Clin Trials 2005; 2: 141–151.

    Article  Google Scholar 

  21. 21

    The Cochrane Collaboration. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011) Higgins JPT GS (ed.), The Cochrane Collaboration: Copenhagen, 2011.

    Google Scholar 

  22. 22

    Zar JH Biostatistical Analysis 2nd edn Englewoods Cliffs New Jersey, USA: Prentice-Hall Inc, 1964.

    Google Scholar 

  23. 23

    Higgins JP, Thompson SG, Deeks JJ, Altman DG . Measuring inconsistency in meta-analyses. BMJ 2003; 327: 557–560.

    Article  Google Scholar 

  24. 24

    The Cochrane Collaboration. Review Manager (RevMan). The Cochrane Collaboration: Copenhagen, 2011.

  25. 25

    Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S et al. Grading quality of evidence and strength of recommendations. BMJ 2004; 328: 1490.

    Article  Google Scholar 

  26. 26

    Birlouez-Aragon I, Saavedra G, Tessier FJ, Galinier A, Ait-Ameur L, Lacoste F et al. A diet based on high-heat-treated foods promotes risk factors for diabetes mellitus and cardiovascular diseases. Am J Clin Nutr 2010; 91: 1220–1226.

    CAS  Article  Google Scholar 

  27. 27

    Cai W, He JC, Zhu L, Peppa M, Lu C, Uribarri J et al. High levels of dietary advanced glycation end products transform low-density lipoprotein into a potent redox-sensitive mitogen-activated protein kinase stimulant in diabetic patients. Circulation 2004; 110: 285–291.

    CAS  Article  Google Scholar 

  28. 28

    Dittrich R, Dragonas C, Kannenkeril D, Hoffmann I, Mueller A, Beckmann MW et al. A diet rich in Maillard reaction products protects LDL against copper induced oxidation ex vivo, a human intervention trial. Food Res Int 2009; 42: 1315–1322.

    CAS  Article  Google Scholar 

  29. 29

    Foerster A, Henle T . Glycation in food and metabolic transit of dietary AGEs (advanced glycation end-products): studies on the urinary excretion of pyrraline. Biochem Soc Trans 2003; 31: 1383–1385.

    CAS  Article  Google Scholar 

  30. 30

    Harcourt BE, Sourris KC, Coughlan MT, Walker KZ, Dougherty SL, Andrikopoulos S et al. Targeted reduction of advanced glycation improves renal function in obesity. Kidney Int 2011; 80: 190–198.

    CAS  Article  Google Scholar 

  31. 31

    Negrean M, Stirban A, Stratmann B, Gawlowski T, Horstmann T, Götting C et al. Effects of low- and high-advanced glycation endproduct meals on macro- and microvascular endothelial function and oxidative stress in patients with type 2 diabetes mellitus. Am J Clin Nutr 2007; 85: 1236–1243.

    CAS  Article  Google Scholar 

  32. 32

    Stirban A, Negrean M, Stratmann B, Götting C, Salomon J, Kleesiek K et al. Adiponectin decreases postprandially following a heat-processed meal in individuals with type 2 diabetes: an effect prevented by benfotiamine and cooking method. Diabetes Care 2007; 30: 2514–2516.

    Article  Google Scholar 

  33. 33

    Stirban A, Negrean M, Götting C, Uribarri J, Gawlowski T, Stratmann B et al. Dietary advanced glycation endproducts and oxidative stress: in vivo effects on endothelial function and adipokines. Ann N Y Acad Sci 2008; 1126: 276–279.

    CAS  Article  Google Scholar 

  34. 34

    Stirban A, Negrean M, Götting C, Stratmann B, Gawlowski T, Mueller-Roesel M et al. Leptin decreases postprandially in people with type 2 diabetes, an effect reduced by the cooking method. Horm Metab Res 2008; 40: 896–900.

    CAS  Article  Google Scholar 

  35. 35

    Schiekofer S, Franke S, Andrassy M, Chen J, Rudofsky G, Schneider JG et al. Postprandial mononuclear NF-kappaB activation is independent of the AGE-content of a single meal. Exp Clin Endocrinol Diabetes 2006; 114: 160–167.

    CAS  Article  Google Scholar 

  36. 36

    Uribarri J, Peppa M, Cai W, Goldberg T, Lu M, He C et al. Restriction of dietary glycotoxins reduces excessive advanced glycation end products in renal failure patients. J Am Soc Nephrol 2003; 14: 728–731.

    CAS  Article  Google Scholar 

  37. 37

    Peppa M, Uribarri J, Cai W, Lu M, Vlassara H et al. Glycoxidation and inflammation in renal failure patients. Am J Kidney Dis 2004; 43: 690–695.

    CAS  Article  Google Scholar 

  38. 38

    Uribarri J, Cai W, Ramdas M, Goodman S, Pyzik R, Chen X et al. Restriction of advanced glycation end products improves insulin resistance in human type 2 diabetes: potential role of AGER1 and SIRT1. Diabetes Care 2011; 34: 1610–1616.

    CAS  Article  Google Scholar 

  39. 39

    Vlassara H, Cai W, Crandall J, Goldberg T, Oberstein R, Dardaine V et al. Inflammatory mediators are induced by dietary glycotoxins, a major risk factor for diabetic angiopathy. Proc Natl Acad Sci USA 2002; 99: 15596–15601.

    CAS  Article  Google Scholar 

  40. 40

    Vlassara H, Cai W, Goodman S, Pyzik R, Yong A, Chen X et al. Protection against loss of innate defenses in adulthood by low advanced glycation end products (AGE) intake: role of the antiinflammatory AGE receptor-1. J Clin Endocrinol Metab 2009; 94: 4483–4491.

    CAS  Article  Google Scholar 

  41. 41

    Gugliucci A, Kotani K, Taing J, Matsuoka Y, Sano Y, Yoshimura M et al. Short-term low calorie diet intervention reduces serum advanced glycation end products in healthy overweight or obese adults. Ann Nutr Metab 2009; 54: 197–201.

    CAS  Article  Google Scholar 

  42. 42

    Pashikanti S, de Alba DR, Boissonneault GA, Cervantes-Laurean D et al. Rutin metabolites: novel inhibitors of nonoxidative advanced glycation end products. Free Radic Biol Med 2010; 48: 656–663.

    CAS  Article  Google Scholar 

  43. 43

    De La Maza MP, Leiva BA, Gattas V, Petermann M, Garrido F, Hirsch S et al. Fluorescent serum and urinary advanced glycoxidation end-products in non- diabetic subjects. Biol Res 2007; 40: 2003–2012.

    Article  Google Scholar 

  44. 44

    Piroddi M, Palazzetti I, Quintaliani G, Pilolli F, Montaldi M, Valentina V et al. Circulating levels and dietary intake of the advanced glycation end-product marker carboxymethyl lysine in chronic kidney disease patients on conservative predialysis therapy: a pilot study. J Ren Nutr 2011; 21: 329–339.

    CAS  Article  Google Scholar 

  45. 45

    Semba RD, Ang A, Talegawkar S, Crasto C, Dalal M, Jardack P et al. Dietary intake associated with serum versus urinary carboxymethyl-lysine, a major advanced glycation endproduct, in adults: the Energetics Study. Eur J Clin Nutr 2012; 66: 3–9.

    CAS  Article  Google Scholar 

  46. 46

    Sourris KC, Forbes JM . Interactions between advanced glycation end-products (AGE) and their receptors in the development and progression of diabetic nephropathy - are these receptors valid therapeutic targets. Curr Drug Targets 2009; 10: 42–50.

    CAS  Article  Google Scholar 

  47. 47

    Brown AA, Hu FB . Dietary modulation of endothelial function: implications for cardiovascular disease. Am J Clin Nutr 2001; 73: 673–686.

    CAS  Article  Google Scholar 

  48. 48

    Lopez-Garcia E, Schulze MB, Fung TT, Meigs JB, Rifai N, Manson JE et al. Major dietary patterns are related to plasma concentrations of markers of inflammation and endothelial dysfunction. Am J Clin Nutr 2004; 80: 1029–1035.

    CAS  Article  Google Scholar 

  49. 49

    Goldberg T, Cai W, Peppa M, Dardaine V, Baliga BS, Uribarri J et al. Advanced glycoxidation end products in commonly consumed foods. J Am Diet Assoc 2004; 104: 1287–1291.

    CAS  Article  Google Scholar 

  50. 50

    Uribarri J, Woodruff S, Goodman S, Cai W, Chen X, Pyzik R et al. Advanced glycation end products in foods and a practical guide to their reduction in the diet. J Am Diet Assoc 2010; 110: 911–916.

    Article  Google Scholar 

  51. 51

    Tessier FJ . The Maillard reaction in the human body. The main discoveries and factors that affect glycation. Pathol Biol 2010; 58: 214–219.

    CAS  Article  Google Scholar 

  52. 52

    Ames JM . Determination of N epsilon-(carboxymethyl)lysine in foods and related systems. Ann N Y Acad Sci 2008; 1126: 20–24.

    CAS  Article  Google Scholar 

  53. 53

    Assar SH, Moloney C, Lima M, Magee R, Ames JM . Determination of Nepsilon-(carboxymethyl)lysine in food systems by ultra performance liquid chromatography-mass spectrometry. Amino Acids 2009; 36: 317–326.

    CAS  Article  Google Scholar 

  54. 54

    Thornalley PJ . Dietary AGEs and ALEs and risk to human health by their interaction with the receptor for advanced glycation endproducts (RAGE)--an introduction. Mol Nutr Food Res 2007; 51: 1107–1110.

    CAS  Article  Google Scholar 

  55. 55

    Heizmann CW . The mechanism by which dietary AGEs are a risk to human health is via their interaction with RAGE: arguing against the motion. Mol Nutr Food Res 2007; 51: 1116–1119.

    CAS  Article  Google Scholar 

  56. 56

    El-Osta A, Brasacchio D, Yao D, Pocai A, Jones PL, Roeder RG et al. Transient high glucose causes persistent epigenetic changes and altered gene expression during subsequent normoglycemia. J Exp Med 2008; 205: 2409–2417.

    CAS  Article  Google Scholar 

  57. 57

    Caballero AE . Metabolic and vascular abnormalities in subjects at risk for type 2 diabetes: the early start of a dangerous situation. Arch Med Res 2005; 36: 241–249.

    CAS  Article  Google Scholar 

  58. 58

    Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol 2011; 64: 401–406.

    Article  Google Scholar 

  59. 59

    Calder PC, Ahluwalia N, Brouns F, Buetler T, Clement K, Cunningham K et al. Dietary factors and low-grade inflammation in relation to overweight and obesity. Br J Nutr 2011; 106 (Suppl 3), S5–S78.

    CAS  Article  Google Scholar 

  60. 60

    Rohrmann S, Zoller D, Hermann S, Linseisen J . Intake of heterocyclic aromatic amines from meat in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Heidelberg cohort. Br J Nutr 2007; 98: 1112–1115.

    CAS  Article  Google Scholar 

Download references


The first author is the recipient of a National Health and Medical Research Council (NHMRC) Postgraduate Public Health Scholarship APP1039709. NJK thanks Mr Brendan Kellow for IT assistance and Dr Andrew Hahne (Latrobe University) for statistical advice. We thank Dr Naiyana Wattanapenpaiboon (Monash University) for providing comments on the manuscript. NJK is supported by a NHMRC Postgraduate Public Health Scholarship APP1039709.

Author information



Corresponding author

Correspondence to N J Kellow.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on European Journal of Clinical Nutrition website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kellow, N., Savige, G. Dietary advanced glycation end-product restriction for the attenuation of insulin resistance, oxidative stress and endothelial dysfunction: a systematic review. Eur J Clin Nutr 67, 239–248 (2013).

Download citation


  • systematic review
  • advanced glycation end product
  • dietary AGE restriction
  • diabetes
  • inflammation
  • humans

Further reading