INCREASED DIETARY INTAKE AND SERUM LEVELS OF ADVANCED GLYCATION END PRODUCTS (AGES) ARE ASSOCIATED WITH DIABETIC MACULAR EDEMA

Abstract

Aim

Diabetic macular oedema (DME) can develop at all stages of diabetic retinopathy, causing visual impairment and blindness. Modern diets are high in advanced glycation end products (dAGEs), derived from processing methods, exerting a pivotal role in promoting diabetic retinopathy risk. In present study, we investigate the relationship between dietary and serum levels of AGEs and DME in type 2 diabetic subjects.

Methods

This case-control study was conducted between July 2018 and February 2019 on 50 case subjects with DME and 40 healthy controls without DM without DME. The sociodemographic characteristics, nutritional status, and anthropometric measurements were evaluated. The advanced glycation end products (AGEs) and receptor for AGEs (sRAGE) levels in serum were analysed.

Results

The AGEs levels of the DME group were higher than in the control group (p <0.05). sRAGE levels were higher in the DME group, but not statistically significant (p >0.05). The dietary intake of AGEs was higher in the DME group (p <0.05). It was found that an increase in neck circumference increased the risk of DME (p <0.001).

Conclusion

A positive correlation was found between DME and AGEs, dAGE, neck circumference, and waist circumference. For the validity of these results, studies, including controlled nutrition interventions, are needed.

INTRODUCTION

Diabetes mellitus (DM) is a chronic metabolic disease that occurs as a result of impairment in the release of insulin, the response of tissues to insulin or both of these factors. The World Health Organization (WHO) predicts that of the global population, around 750 million will have DM by 2025 (1). The Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR), conducted in the United States, reported a prevalence of diabetic retinopathy (DR) in 71% and 47% of type 1 DM and type 2 DM patients, respectively (2, 3).

DR is the leading cause of vision loss and the most common cause of blindness in diabetic patients. In 2015, it was reported that 2.6 million people were visually impaired because of DR(4-6). DR typically affects working-age adults and thus inflicts indirect costs due to lost productivity in addition to direct costs on the healthcare system (4, 7). Diabetic macular edema (DME) can be seen at any stage of diabetic retinopathy. In WESDR, 29% of patients with type 1 diabetes developed DME over 25 years; in another study by the same team, DME developed in 25.4% of type 2 DM cases being treated with insulin and in 13.9% of those who were not (2, 3).

Poorly managed diabetes is more likely to lead to DR than well-treated diabetes. It has been noted that improving glycaemic control through dietary changes can be beneficial in preventing DR. Studies have shown that hyperglycaemia and diabetes duration are the strongest indicators of retinopathy (8-10).

Nutrients can also play a central role in DR patients’ resistance to conventional medical treatments. Recently nutrition-based approaches have gained momentum in various clinical conditions. As DR is a nutritionally responsive disorder, the defensive role of nutrition in daunting DR deserves the spotlight. But probably due to the byzantine interplay existing between nutrients and DR, the worth of nutrition in curbing this vision-threatening disorder remains silent. A nutritional therapy, if adopted in the initial stages, can be an anodyne option proving effective, inexpensive, and readily available in halting DR onset or progression. Nutrition-based approaches can provide superior efficacy and propose a complementary solution to the clinically-unmet requirement for preventing DR (11).

Hyperglycaemia can lead to DR in patients with DM via many pathways, including the polyol pathway, non-enzymatic protein glycation, the activation of protein kinase C, the activation of the hexosamine pathway, the formation of reactive oxygen species (ROS) and the induction of hypoxia-inducible factor that responds to decreases in available oxygen in the cellular environment, or hypoxia (8-10). Of these, non-enzymatic protein glycation leads to the most accelerated accumulation of advanced glycation end products (AGEs). In addition to the loss of retinal capillary pericytes, it also leads to inflammation, oxidative stress, and the activation of vascular endothelial growth factor (VEGF) (8-10).

Retinal pericytes accumulate AGEs during diabetes, which would be expected to have a detrimental influence on pericyte survival and function. Indeed, we have previously shown that the AGE-RAGE interaction elicits ROS generation in cultured retinal pericytes, thereby inducing apoptotic cell death of pericytes. AGEs induced nuclear factor-κB (NF-κB) activation, decreased the ratio of Bcl-2/Bax, and subsequently increased the activity of caspase-3, a key enzyme in the execution of apoptosis of pericytes. Moreover, AGEs up-regulated RAGE mRNA levels in pericytes through the intracellular ROS generation. The fact that dietary AGEs contribute to the body’s AGE pool, therefore, becomes important for DME (12).

The AGE receptor (RAGE) is a multi-band signal transduction receptor of the immunoglobulin superfamily of cell surface molecules. RAGE, a heterogeneous product group produced by non-enzymatic pathways from the glycation and oxidation of protein, lipid, and DNA amino groups, was first described as a receptor for AGEs (13). Decreased sRAGE levels can be used as a marker for disease status.

Today, the application of cooking and preserving methods to increase the flavour of food, lengthen shelf life, and eliminate foodborne diseases is common. AGEs are formed as a result of Maillard reactions between carbohydrates and proteins when found in high amounts in foods. Dietary AGEs are pathogenic compounds that can lead to the occurrence and progression of many chronic diseases. Recent studies involving humans and experimental animals have found dietary AGEs to be absorbed and contribute significantly to the body’s AGE pool (14-16).

In the present study, the serum levels of AGEs, sRAGE, dietary AGEs, food and nutrient intake; and some anthropometric measurement results of individuals with type 2 DM were determined. This cross-sectional study aimed to evaluate the associations between DME and some nutritional factors, which can be applied to arrest the pathogenesis of DME and sustain good vision among diabetic subjects.

METHODS

This cross-sectional study was conducted on 90 type 2 DM patients and approved by University Non-Interventional Clinical Research Ethics Committee on June 12, 2018, with project number GO 18/562 and decision number GO 15/628-20. The reporting of this work is compliant with STROBE guidelines. This case-control study was conducted between July 2018 and February 2019 on 50 case subjects with DME and 40 healthy controls with DM but without DME.

While benefiting from the results of the previous studies, the sample size of the research was type 1 error level α = 0.05 and type 2 level β = 0.20. The power of the test was taken as 1–β = 0.80, and the power analysis was statistically calculated using NCCS PAS 11 program. Ninety patients with type 2 DM were divided into two groups: those with and without DME. All patients were ≥30 years old and were diagnosed with type 2 DM. Fifty patients with DME were evaluated as the case group and 40 patients without DME were evaluated as the control group. All DME patients who applied to the hospital for the study were enrolled. Case patients with any disease other than DR that may affect the retina, or with any disease that may affect the retina for the control patients; with corneal, lens or vitreous opacification preventing optical coherence tomography (OCT) application, with any systemic disease other than type 2 DM and hypertension; or with a special diet or nutrient intake were excluded. Written informed consent was obtained from all participants. After a detailed systemic and ophthalmological history was taken from all participants, best-corrected visual acuity (BCVA) was measured using an ETDRS chart. Dilated fundus examination was performed with a 90 D lens after anterior segment examination with a biomicroscope. Spectral-domain OCT (Spectralis OCT, Heidelberg Engineering, Heidelberg, Germany) was applied for the diagnosis of DME.

The body weight of each participant was measured using a calibrated electronic scale with an accuracy of 0.1 kg (TANITA HD 366) and conducted when the patient was fasted, wearing lightweight clothes and no shoes. Body height was measured by a researcher (S.A.) using a tape measure when each participant was standing upright in the Frankfurt Plane position (i.e., the ear canal and the lower margin of the orbit are horizontal while facing forward). BMI was calculated using the formula body weight (kg)/body length (m²) and categorized based on the WHO classification. According to this classification, a BMI of <18.5 kg/m² is thin, 18.5–24.9 kg/m² is normal, 25.0–29.9 kg/m² is overweight, and ≥30 kg/m² is obese. Waist circumference was also measured using a tape measure by circling the waist at the midpoint between the lowest rib and the umbilicus. If the waist circumference is found to be ≥88 cm for women and ≥102 cm for men, this is indicative of a high risk of developing chronic diseases (20). Hip circumference was subsequently measured using a tape measure at the highest point of the hip while standing at the left side of the individual. If the waist-to-hip ratio is ≥0.85 for women and ≥0.90 for men, the risk of developing chronic diseases increases.

The dietary intake of participants was assessed using a validated quantitative food frequency questionnaire (QFFQ) and a three-day food record form (17). Meanwhile, to assess dietary AGEs based on QFFQ, each food’s contribution to dAGEs intake was calculated based on the Advance Glycation End Products in Foods Table published by Uribarri et al. (18), which comprises data on 549 food items. Standardized food recipes for Turkey and the Nutrition Information System (BEBIS) program, which is a food composition database for nutrient estimation, were used to determine the average daily energy and nutrient intake for each participant. These values were subsequently compared with the recommended daily allowance values to determine the status of meeting energy and nutrient requirements. After that, the percentages meeting the requirements were calculated.

We evaluated the physical activity status with a 24-hour Physical Activity Assessment Questionnaire (24-h PAAQ) (19) which were developed for people working (range 18-65 years) in different professions. The questionnaires were performed by interview technique. They consist of different activity groups which are work, non-work (leisure time and sleeping) and sociodemographic characteristics. The amount of MET consumed per day (MET/day) was calculated for all questionnaires.

The enzyme-labelled immune “Enzyme-Linked Immunosorbent Assay (ELISA)” test was performed by biologists using a Biotech 800TS device. Carboxymethyl lysine (CML) was considered to be the AGE parameter, being as most easily detected and abundant type of AGE in humans. To analyze the samples, Human CML Elisa Kit 96 tests for serum AGE and Human RAGE Elisa Kit 96 test kits for serum RAGE were used.

For statistical evaluation of the data obtained in the study, the Statistical Package for Social Science 22.0 program (IBM, USA) was used. Normally distributed data were analyzed with parametric statistical tests (independent two-sample t-test and Pearson’s correlation), whereas non-normally distributed data were analyzed using non-parametric statistical tests (Pearson’s Chi-square, Mann-Whitney U and Spearman’s correlation). Odds ratios (OR) and 95% confidence intervals (CI) were calculated using a univariate logistic regression analysis independently for each criterion (food group, BMI, adjusted age, gender, diabetes duration and glucose) before combining them in multivariate logistic regression. The OR and 95% CI values were then used to determine the risk factors for developing DME. Furthermore, a stepwise multivariate regression analysis was also performed, with strict inclusions of the set with p < 0.05 in the model.

RESULTS

General characteristics, anthropometric measurements of the groups and statistical analysis results are presented in Table 1.

Table 1.

General anthropometric characteristics and socio-demographic data

Sociodemographic characteristics	Case (n=50)			Control (n=40)		Total (n=90)		P
	S	%		S	%	S	%
Gender
Female	30	60.0		26	65.0	56	62.2
Male	20	40.0		14	35.0	34	37.8	0.627a
Total	50	100.0		40	100.0	90	100.0
Age (mean (years)) (X±SD) (min-max)	62.6±8.4 (36.0-82.0)			58.7±7.8 (44.0-82.0)		60.9±8.4 (36.0-82.0)		0.024b
Total education period (years) (min-max) Median (IQR)a	(0-15) 5 (6)			(0-13) 5 (6)		0-15 5 (6)		0.852c
Occupation
Worker	5		10.0	8	20.0	13	14.4
Non-worker	45		90.0	32	80.0	77	85.6	0.182c
Total	50		100.0	40	100.0	90	100.0
Sports/Exercise status
No	43		86.0	23	57.5	66	73.3
Yes	7		14.0	17	42.5	24	26.7	0.002*
Total	50		100.0	40	100.0	90	100.0
Diabetes duration (min-max) Median (IQR) (years)	3-40 15 (10.8)			2-23 9 (8.8)		2-40 12 (11.3)		0.001***
Body mass index (BMI X± SD) (kg/m2)	33.1±7.2			31.6±5.8		32.4±6.6		0.477***
Female BMI	35.5±7.5			33.0±6.4		34.3±7.1		0.257***
Male BMI	29.4±5.0			29.2±3.5		29.3±4.4		0.796***
Waist circumference (WC; _X± SD) (cm)	106.1±12.6			104.3±7.9		105.3±10.8		0.804***
Female WC	110.6±13.0			106.5±7.9		108.7±11.0		0.263***
Male WC	99.4±8.5			100.2±6.4		99.7±7.6		0.691***
Waist-to-hip ratio (W/H X± SD) (cm)	0.96±0.05			0.94±0.04		0.95±0.05		0.179***
Female W/H	0.96±0.05			0.93±0.05		0.95±0.05		0.204***
Male W/H	0.97±0.05			0.95±0.03		0.96±0.04		0.592***
Neck circumferences (NC. cm)	41.7±4.1			38.4±3.8		40.2±4.3		0.000***
Female NC	42.4±4.0			38.3±4.1		40.5±4.5		0.000***
Male NC	40.6±3.9			38.4±3.4		39.7±3.8		0.071***

aIQR; Interquartile range, *Pearson chi-squared test, **Independent two-sample t-test, ***Mann-Whitney U test (p<0.05).

When the gender distribution of the participants was analyzed, it was found that 62.2% of the participants were female, and 37.8% were male. There was a statistical difference between the average age of the groups (p <0.05). The difference in diabetes duration was examined, and a statistically significant difference was found between the case and the control groups in terms of the duration of diabetes (p <0.001). These results suggest that long-term DM during adulthood may lead to DME in later life.

When the average neck circumferences were examined, the neck circumference of the case group was found to be statistically significantly higher than that of the control group (p <0.001). The mean neck circumference of the case group was 41.7±4.1 cm, and the mean neck circumference of the control group was 38.4±3.8 cm.

Biochemical parameters, dietary AGEs intakes of the participants and statistical analysis results are presented in Table 2.

Table 2.

Biochemical parameters and dietary intake levels of individuals (min-max) median (IQR) and `X± SD

	Case (n=50)	Control (n=40)	Total (n=90)	p
Serum AGEs (ng/mL) (CML)	162.4-1565.0 451.7 (425.0)	150.0-1474.1 422.5 (120.0)	150-1565 425.4	0.023
sRAGE (pg/mL)	79.6-743.2 186.2 (111.0)	67.2-727.9 159.7 (62.6)	67.2-743.2 176.5	0.169
AGEs/sRAGE	0.51-9.52 2.69 (3.77)	0.83-13.18 2.42 (1.33)	0.51-13.18 2.53	0.342
HbA1c (%)	7.4±1.0	7.0±0.9	7.2±1.0	0.042
Fasting blood glucose (mg/dL)	158.1±66.1	150.0±43.3	154.5±56.9	0.484
Dietary Total AGE intake (kU/g) (CML)	10984.9±3767.5	8626.0±3043.9	9936.5±3642.0	0.005
Dietary Glycaemic index	65.6±12.3	61.8±12.9	63.9±12.6	0.122
Serum Lipid profile (mg/dL)
Total Cholesterol	180.3±30.4	183.4±29.9	181.7±30.1	0.628
HDL-C	44.6±11.2	46.5±11.6	45.5±11.4	0.429
LDL-C	116.8±26.3	117.1±34.8	116.9±30.2	0.968
TG	151.6±70.1	146.6±75.4	149.3±72.1	0.746

IQR; Interquartile range, AGEs; advanced glycation end products, sRAGE; advanced glycation end products receptor Mann-Whitney U test (p<0.05).

Although the fasting blood glucose levels of those in the case group were high, the difference was not statistically significant (p >0.05). The HbA1c levels of the case group were significantly higher than in the control group (p <0.05).

No statistically significant difference was found between the total cholesterol levels of the groups (p>0.05). There was also no statistically significant difference between the HDL and LDL levels of the groups (p >0.05).

The serum AGEs, sRAGE, and Dietary AGEs intakes of the participants are shown in Table 2. The Serum AGEs levels in the case group (a group with DME) were higher than in the control group. (p <0.05). There was no statistically significant difference in sRAGE values of the groups (p >0.05).

When the dietary AGEs consumption of the groups was examined, the diet AGE consumption of the case group was found to be statistically significantly higher than that of the case group (p <0.001). When the AGE/sRAGE ratio, which is thought to be an indicator of the DME, was analyzed, no statistically significant difference was identified between the groups (p >0.05).

Some risk factors were evaluated between the case and control groups using a statistical logistic regression analyzis, adjusted for age, and the results of the risk of DME are presented in Table 3.

Table 3.

Univariate Binary Logistic Regression Analysis for DME versus No DME, Adjusted for Age, Gender, Diabetes Duration and Glucose

	OR	%95CI	p
Regular exercises	0.26	0.02-1.83	0.011
Insulin treatment duration	1.15	0.99-1.33	0.063
Neck circumference	1.24	1.08-1.41	0.001
Waist circumference	1.06	1.01-1.13	0.027
Dietary AGEs	1.24	1.08-1.42	0.03
Serum AGEs	1.22	1.03-1.43	0.021
Serum sRAGE	1.16	0.81-1.67	0.424
Serum calcium	0.54	0.20-1.48	0.23
Dietary Glycaemic index	1.03	0.99-1.07	0.088
Dietary Fat (%)	1.05	1.00-1.10	0.037
Vitamin K	0.99	0.99-1.0	0.029
Vitamin B2	0.16	0.04-0.57	0.005
Vitamin B6	0.21	0.07-0.65	0.007
Folate	0.99	0.99-1.0	0.000
Vitamin C	0.99	0.99-1.0	0.013
Calcium	0.99	0.99-1.0	0.003
Saturated fatty acid	1.13	1.06-1.21	0.000
Short-chain fatty acids	3.70	1.41-9.74	0.008
Medium-chain fatty acids	5.43	1.58-18.63	0.007

DME; Diabetic macular edema Note. Boldface data show significant difference. OR: odds ratio; CI: confidence interval.

A longer duration of insulin treatment (OR: 1.15, p = 0.063) and a higher glycaemic index (OR: 1.03, p = 0.088) were not statistically significantly associated with DME risk. When the parameters that increase the risk of the disease are examined, increased duration of diabetes and increased neck circumference were increased the risk of DME (p <0.05). Higher energy diet consumption increased the risk of DME. (OR: 1.05, p <0.05). An increase in saturated fatty acid intake increases the risk of DME 1.13 times (p <0.001). It was found that an increase in the consumption of medium-chain fatty acids increased the risk of DME 5.43 times (p <0.05). It was also found that increased waist circumference also increased the risk of disease (OR: 1.06, p <0.05).

A one-unit increase in serum AGEs levels was found to increase the risk of DME 1.22 times and was found to be statistically significant (p <0.05). Although the increase in sRAGE levels was found to increase the risk of disease 1.16 times, this result was not statistically significant (p >0.05).

Daily energy and nutrient intakes of individuals are presented in Table 4. Fat and saturated fat intakes of individuals in the case group were higher than those in the control group (p <0.05). Fiber, vitamin C, folic acid and calcium intakes were found to be lower in the case group than in the control group (p <0.05). When MUFA and PUFA intakes were examined, no statistically significant difference was found between the two groups (p >0.05).

Table 4.

Daily energy and nutrient intakes of individuals (min-max) and `X± SD

Energy and Nutrients	Case (n=50)	Control (n=40)	Total (n=90)	p
Energy (kcal)	1472±467.4 (701.3-2700.5)	1405±384.7 (750.0-2651.0)	1442.8±431.6 (701.3-2700.5)	0.500*
Protein (g)	62.6±26.0 (24.1-135.1)	68.6±29.2 (30.3-179.6)	65.3±27.5 (24.1-179.6)	0.249*
CHO (g)	159.5±61.8 (64.3-371.1)	156.9±42.9 (69.5-257.9)	158.3±53.9 (64.3-371.1)	0.609*
Fat (g)	63.4±24.2 (17.8-137.6)	53.8±19.3 (20.9-107.4)	59.1±22.5 (17.8-137.6)	0.042*
Saturated fatty acid(g)	21.4±8.8 (5.9-47.2)	14.9±6.7 (6.3-40.1)	18.5±8.6 (5.9-47.2)	<0.001*
MUFA(g)	19.6±8.6 (4.4-49.8)	17.7±7.5 (5.2-36.3)	18.8±8.1 (4.4-49.8)	0.247*
PUFA(g)	16.7±8.2 (4.3-43.3)	16.0±6.7 (5.2-30.2)	16.4±7.6 (4.3-43.3)	0.658
Omega 3 (g)	1.0±0.7 (0.3-3.6)	1.2±0.9 (0.5-4.5)	1.1±0.8 (0.3-4.5)	0.376*
Omega 6 (g)	15.5±7.8 (3.2-39.3)	14.8±6.2 (4.5-28.3)	15.2±7.1 (3.2-39.3)	0.628
Cholesterol (mg)	265.4±194.1 (17.2-821.4)	241.5±158.9 (14.7-789.4)	254.8±178.8 (14.7-821.4)	0.523*
Fiber (g)	20.3±107 (6.7-61.6)	22.6±8.4 (9.4-41.3)	21.3±9.8 (6.7-61.6)	0.069*
Vitamin A (mcg)	1015.3±967.7 (192.2-5516.4)	1070.8±946.5 (297.9-5202.7)	1039.9±953.4 (192.2-5516.4)	0.751*
Carotene (mcg)	4.2±5.7 (0.3-31.1)	5.2±5.8 (0.7-31.1)	4.7±5.7 (0.3-31.1)	0.063*
Lutein + Zeaxanthin	-	5.5±14.0 (0.0-48.0)	2.5±9.7 (0.0-48.0)	0.005*
Vitamin E (mg)	17.2±8.9 (2.6-41.5)	17.8±7.5 (6.4-34.5)	17.5±8.3 (2.6-41.5)	0.543*
Vitamin C (mg)	106.5±78.7 (4.3-376.9)	144.2±71.2 (59.1-323.5)	123.2±77.4 (4.3-376.9)	0.003*
Thiamine (mg)	0.8±0.3 (0.3-1.6)	0.9±0.3 (0.4-1.7)	0.8±0.3 (0.3-1.7)	0.252*
Riboflavin (mg)	1.1±0.4 (0.4-2.2)	1.4±0.4 (0.6-3.0)	1.2±0.4 (0.4-3.0)	0.001
Vitamin B6 (mg)	1.2±0.4 (0.3-2.3)	1.5±0.5 (0.9-3.1)	1.3±0.5 (0.3-3.1)	0.006
Vitamin B12 (mcg)	4.0±3.6 (0.2-19.9)	5.6±4.7 (0.6-19.1)	4.7±4.2 (0.2-19.9)	0.079*
Folic acid (mcg)	299.1±118.8 (93.9-700.0)	405.3±132.1 (194.2-686.6)	346.3±135.1 (93.9-700.0)	<0.001
Calcium (mg)	573.1±211.9 (300.0-1114.4)	733.4±261.1 (385.9-1640.7)	644.4±247.0 (300.0-1640.7)	0.002
Magnesium (mg)	247.5±94.1 (88.0-491.5)	264.2±84.2 (121.8-471.2)	254.9±89.7 (88.0-491.5)	0.299*

DISCUSSION

DR is a common microvascular complication in patients with DM with debilitating effects on visual acuity and may eventually lead to blindness (20). The longer the duration of diabetes, the higher the prevalence of DR (21). DME, which is a complication of DR, doubles the rate of cardiovascular morbidity and mortality in such patients (22).

Both Klein et al. and Varma et al. identified positive relationships between a more prolonged duration of diabetes and DME (23, 24). Similarly, prolonged diabetes duration was found to be a risk factor for DME in the present study (p <0.001). Furthermore, the logistic regression analyzis found that a one-unit increase in the duration of diabetes increased the risk of DME 1.12 times (p <0.05).

It is well known that physical activity reduces blood pressure and serum lipids, including triglycerides, and improves glycaemic control (25). In the present study, a negative relationship was found between physical activity and DME (p=0.002). Physical activity, in addition to other advantages, is critical in preventing type 2 DM. It can help you maintain weight loss, enhance heart health, and fight insulin resistance by increasing the body’s insulin response.

Neck circumference is an equivalent indicator of upper body subcutaneous fat distribution and has been closely linked to various metabolic risk factors (26). In the present study, a statistically significant positive relationship was found between DME and neck circumference (p <0.001). Neck circumference may be considered as a predictive parameter in the diagnosis and treatment of DME. Also, according to our logistic regression analyzis results, it was found that an increase in the neck circumference increased the disease risk by 1.24 times (p < 0.001).

Waist circumference and waist-hip ratio are useful markers of visceral adipose tissue.DR has also been associated with visceral adiposity (27). In two studies examining the relationship between waist circumference and waist-hip ratio and DR/DME, a relationship was found between waist circumference and waist/hip ratio and DME in females but not in males (OR, 3.49 [95% CI, 1.50-8.10] and OR, 2.68 [95% CI, 1.28-5.62]) (27, 28). In the present study, we found no relationship between waist circumference, hip circumference, waist/hip ratio distribution, and DME in both case and control groups or the gender subgroups (p >0.05).

High levels of AGEs are believed to play a causal role in DR (29). While there have been many reports measuring several disease-defined AGE molecules, most of the studies have used specific antibodies for CML, and have found a relationship between diabetic retinopathy and serum AGEs (30-32). In the present study, a serum CML analyzis was performed using the ELISA method in which the DME group’s AGEs levels were found to be statistically significantly higher than those of the control group (p = 0.023).

Advanced glycation products (AGEs) and their receptors (RAGE) play a significant role in the development of complications in diabetes (33-35). One study identified sRAGE as an essential biomarker in RAGE-mediated pathogenesis (36). In the study by Nakamura et al. (37), the sRAGE level was found to be higher in type 2 DM patients than non-DM patients. In the present study, sRAGE was found to be higher in the DME group, but with no statistically significant difference between the groups (p >0.05). Also, in the present study, although a one-unit increase in sRAGE levels was found to increase the risk of DME 1.16 times, this finding was not statistically significant (p >0.05). The difference in results may be since patients with diabetic retinopathy were not included in the current study.

The role of dietary AGEs in the development of chronic complications in diabetes is controversial. That said, recent studies have reported that the long-term reduction of heat-enriched dAGEs reduces insulin resistance, improves wound healing impairment, and prevents diabetes and the development of related vascular and kidney complications in animal models (38, 39). In the present study, concurring with the findings of previous studies in the literature, dAGEs intake was found to be statistically significantly higher in the DME group (p = 0.005). Furthermore, our logistic regression analysis results indicate that an increase in dAGEs intake increases the risk of DME 1.24 times (p <0.05).

Sasaki et al. (40) observed no relationship between monounsaturated fatty acids (MUFA) and retinopathy, while Alcubierre et al. (41) found MUFA and oleic acid to be inversely related to retinopathy. There is evidence suggesting that polyunsaturated fatty acids (PUFA) intake is beneficial in preventing retinopathy (40, 42, 43), although no such association has been observed in any study (41). n-3 PUFA supplementation has been shown to reduce the number of retinal cellular capillaries associated with diabetes, as well as inflammatory markers in the retina of diabetic animal models (44, 45). However, it may be necessary to examine the effects of PUFAs separately, as their effects may differ within the group, and even within subgroups (46). In our study, the total fat, saturated fatty acid consumption and percentage of dietary energy from fat in the DME group were found to be higher than in the control group, and this difference was statistically significant (p <0.05). When the total cholesterol consumption was evaluated with MUFA, PUFA, and omega-6 fatty acids, it was found to be higher in the DME group, although there was no statistically significant difference when compared to the consumption in the control group (p >0.05). The n-3 fatty acid consumption of the control group was found to be higher than the DME group, but this difference was not statistically significant (p >0.05).

Vitamin C intake was associated with a reduced risk of retinopathy in a prospective study (47), although no such negative relationship was reported in two cross-sectional studies (48, 49). In the cross-sectional National Health and Nutrition Examination Study (NHANES III), an inverse relationship was observed between serum vitamin C and retinopathy, although it was not statistically significant and was noted only after the exclusion of vitamin C supplement users (50). In the present study, the consumption of vitamin C (144.2±71.2 mg) in the control group was found to be higher than that of the DME group (106.5±78.7 mg), and the difference was statistically significant (p = 0.003). Also, in the present study, the intake of B2, B6, and folic acid was found to be higher in the control group than in the DME group, and these differences were statistically significant (p <0.05).

The soluble receptor for advanced glycation end products (sRAGE) and AGEs has been associated with risks of cardiovascular disease (51). In the present study, the AGE/sRAGE ratios of patients who were non-DME and DME were evaluated, and no significant statistical correlation was found between AGE/sRAGE and DME (p = 0.342).

In the present study, the dAGEs intake of the DME group was 10,985±3,768 kU/day, and the dAGEs intake of the control group was 8,626±3,044 kU/day. Similar results have been reported by other investigators (52, 53). Macías-Cervantes et al. (53) reported an average 14,311 kU/day intake of AGEs in the Mexican population aged 30-55 years. The reason why the dAGEs intake of the participants in the present study was below most studies in the literature is that our participants had a higher average age and were patients with diabetes. Most of the large-series studies of dAGEs intake in literature have been conducted with healthy people and middle-aged people.

The findings of this study have to be seen in the light of some limitations. Comparison of study results could not be conducted efficiently in the study because of a lack of previous research studies on the DME and dietary AGEs relationship. Since it is a single-centre study, the diet and nutritional habits at the site of the study may create limitations in generalising the results of the study.

DME is a multifactorial disorder, although nutrition is one of the most modifiable risk factors. Our study demonstrated that a high dietary AGEs intake, a high level of AGEs and HbA1c, and longer diabetes duration might act as a potential risk factor for DME development. We could find no significant relationship between sRAGE and DME, the soluble receptor of AGEs. With the increasing prevalence of DM, effort must be spent to reduce the risk of DME development through dietary blood sugar control. Low dietary AGEs intake, especially in patients with diabetes, can provide additional benefits.

Conflict of interest

The authors declare that they have no conflict of interest.