Serum advanced glycation endproducts are associated with left ventricular dysfunction in normal glucose metabolism but not in type 2 diabetes: The Hoorn Study

Abstract

Objective:

To investigate whether serum advanced glycation endproducts are associated with left ventricular systolic and diastolic function in participants with normal glucose metabolism, impaired glucose metabolism and type 2 diabetes mellitus.

Methods:

Participants from a cross-sectional, population-based study (n = 280 with normal glucose metabolism, n = 171 with impaired glucose metabolism, n = 242 with type 2 diabetes mellitus) underwent echocardiography. Serum protein-bound advanced glycation endproducts [i.e. Nε-(carboxymethyl)lysine, pentosidine and Nε-(carboxyethyl)lysine] were measured. Linear regression analyses were used and stratified according to glucose metabolism status.

Results:

In normal glucose metabolism, higher Nε-(carboxymethyl)lysine and pentosidine levels were associated with worse diastolic function (left atrial volume index and left atrial volume × left ventricular mass index product term) and higher Nε-(carboxymethyl)lysine and Nε-(carboxyethyl)lysine levels with worse systolic function (ejection fraction). In impaired glucose metabolism, a similar pattern emerged, though less consistent. In type 2 diabetes mellitus, these associations were non-existent for diastolic function or even reversed for systolic function.

Conclusion:

This suggests that serum advanced glycation endproducts are associated with impaired left ventricular function in normal glucose metabolism, but that with deteriorating glucose metabolism status, serum advanced glycation endproducts may not mirror heart failure risk.

Keywords

Advanced glycation endproducts heart failure echocardiography type 2 diabetes mellitus deteriorating glucose metabolism status

Introduction

Type 2 diabetes mellitus (T2DM) is associated with an increased risk of heart failure.¹ This increased risk is only partially understood as it is not fully explained by traditional cardiovascular disease (CVD) risk factors.¹

It has been suggested that advanced glycation endproducts (AGEs) play an important role in the pathobiology of heart failure.^2,3 For example, AGEs may impair both diastolic and systolic left ventricular (LV) function directly, through extracellular matrix protein cross-linkage and/or altered calcium handling, and/or indirectly, through the interaction with cardiac AGE receptors, which in turn lead to low-grade inflammation, increased oxidative stress and altered gene expression.^2–4 Consequently, AGEs may also unfavourably affect the process of LV remodeling.^2–4 The impact of AGEs on LV function may be of particular importance in T2DM, as T2DM is characterized by chronic hyperglycaemia and hyperlipidaemia, which in turn can induce the formation of AGEs, as well as oxidative stress and low-grade inflammation, which can further enhance the formation of AGEs.^2,4–6

However, previous studies on the association between circulating AGEs and LV structure and function showed mixed results. For example, some studies,^7–9 but not all,¹⁰ showed that higher circulating AGEs were associated with impaired diastolic function. Such mixed results were also seen for systolic function. For example, Berg et al.⁸ reported positive associations of circulating AGEs with systolic function, whereas such associations were absent in a study by Steine et al.¹¹ In addition, only a few studies reported on the association between circulating AGEs and indices of LV structure.^8,10 These studies were, in addition, small (i.e. n < 60),^7–11 done in selected populations (e.g. patients undergoing coronary bypass surgery,⁷ patients with type 1 diabetes^8,9,11 or with end-stage renal disease¹⁰) and did not investigate the role of deteriorating glucose metabolism on the association between circulating AGEs and LV function.^8–11

Therefore, we investigated, in a large population-based cohort stratified according to glucose metabolism status (GMS), the associations between the serum AGEs, Nε-(carboxymethyl)lysine (CML), pentosidine and Nε-(carboxyethyl)lysine (CEL) and LV function and structure. We hypothesized that serum AGEs would be associated with impaired LV function and unfavourably altered LV structure. In addition, we hypothesized that any such association would be stronger in individuals with impaired glucose metabolism (IGM) and T2DM.

Methods

Study population

For the current study, we used data from the 2000 Hoorn Study. The Hoorn study, which started in 1989, is a population-based cohort study of GMS in relation to CVD risk factors.¹² In brief, 2484 men and women, aged 50–75 years, from the population register of the medium-sized Dutch town of Hoorn participated in the baseline examination. In 1996–1998, 1513 (73%) of all surviving participants agreed to participate in a follow-up visit. In 2000, all those who were diagnosed as having diabetes during the previous examinations (n = 176) and random samples of individuals with normal glucose metabolism (NGM) (n = 705) and IGM (n = 193) were invited, of whom 648 (60%) participated. To increase the number of individuals with T2DM, 188 individuals with T2DM from the Hoorn Screening study were added.¹³ The total study population, therefore, consisted of 836 individuals.¹⁴ The study complied with the Declaration of Helsinki. The Medical Ethical Review Committee of the VU University Medical Center, Amsterdam, has approved the research protocol, and written informed consent has been obtained from all participants.

Measurement of AGEs

AGEs were measured in serum samples, obtained from fasting venous blood, which were stored at −80°C until analysis. Protein-bound CML, CEL and lysine were measured by means of ultra-performance liquid chromatography tandem mass spectrometry, as described in detail elsewhere.¹⁵ Intra- and interassay coefficients of variation were 4.8% and 7.0% for CML, 5.0% and 9.7% for CEL and 2.7% and 4.9% for lysine, respectively.¹⁵ Protein-bound pentosidine was measured using high-performance liquid chromatography with fluorescent detection, as described in detail elsewhere.¹⁶ Intra- and interassay coefficients of variation for pentosidine were 6.5% and 3.1%, respectively.¹⁶ Concentrations of protein-bound CML, CEL and pentosidine were adjusted for levels of lysine and expressed as nmol/mmol lysine.

Echocardiography

Echocardiography was performed according to a standardized protocol consisting of two-dimensional and M-mode assessments as previously described.¹⁷ A Hewlett-Packard SONOS 5500 echocardiography system (2–4 MHz transducer, Andover, MA, USA) was used. Measurements of LV structure [as determined by interventricular septum thickness, posterior wall thickness, LV end-diastolic diameter, relative wall thickness and left ventricular mass index (LVMI)] were performed in M-mode and determined according to the American Society of Echocardiography 2005 recommendations.¹⁸ Systolic function was defined with the use of monoplane Simpson’s LV ejection fraction, calculated from the LV systolic and diastolic volumes in the apical four chamber view. For diastolic function, two markers were used: left atrial (LA) volume index and the LA volume × LVMI product term.^19,20

Other measurements

Individuals’ GMS was defined according to the 1999 WHO criteria²¹ with an oral glucose tolerance test as previously described.¹⁴ Study covariates included age, sex, GMS, hypertension (defined as systolic/diastolic blood pressure greater than 140/90 mmHg and/or use of blood pressure-lowering medication²²), total cholesterol–high-density lipoprotein cholesterol ratio, triglycerides, use of lipid-modifying medication, prior CVD, current smoking, waist circumference, estimated glomerular filtration rate (eGFR),²³ homeostasis assessment model for insulin resistance (HOMA-IR),²⁴ use of insulin therapy and use of angiotensin-converting enzyme (ACE) inhibitors or other renin angiotensin aldosterone system (RAAS)-modifying medication, all determined as previously described.^12,13

Statistical analysis

All analyses were performed with the statistical software package SPSS version 21.0 (SPSS IBM Corporation, Armonk, NY, USA). Descriptive statistics are presented as means ± standard deviations, or in case of a skewed distribution as median (interquartile range), or frequencies (percentages). Variables with a skewed distribution (i.e. CML, CEL, pentosidine, triglycerides, LA volume index, LA volume × LVMI product term) were log-transformed in order to meet normality criteria. For reasons of direct comparison, the individual AGEs Z-scores [individual value − population mean)/(population standard deviation] were calculated for CML, CEL and pentosidine. For descriptive purposes and legibility, LA volume index and LA volume × LVMI product term values were multiplied by −1 so that lower values indicate worse diastolic function.

We used linear regression analyses to investigate the associations between Z-scores of AGEs and LV structure and function. Adjustments were made first for age and sex (model 1). The results were then additionally adjusted prior to CVD, hypertension and/or the use of anti-hypertensive medication, total cholesterol–high-density lipoprotein cholesterol ratio, triglycerides, the use of lipid-modifying medication, current smoking, eGFR and waist circumference (model 2). Initial analyses showed a statistically significant interaction between AGEs and GMS in their association with LV function, and therefore, the analyses were stratified according to GMS. p < 0.05 was considered statistically significant, except for the interaction analyses, where p < 0.10 was used.

Results

Study population

For the present analyses, individuals were excluded with missing data on serum AGEs (n = 73), GMS (n = 5) or other covariates (n = 54). In the remaining 704 individuals, full data to determine LV dimension, LV systolic function and LV diastolic function were available in 693, 649 and 672 individuals, respectively. The final study population (n = 693) consisted of 280 individuals with NGM, 171 with IGM and 242 with T2DM. Individuals with missing data more frequently had diabetes and hypertension, had a higher waist circumference and higher levels of triglycerides (data not shown).

Characteristics

The characteristics of the study population stratified according to tertiles of diastolic function (i.e. tertiles of LA volume index) are given in Tables 1 and 2. With deteriorating diastolic function, participants were, in general, older, suffered more frequently from hypertension and used more anti-hypertensive medication (including ACE inhibitors). In addition, these participants more often had T2DM, prior CVD, albuminuria and lower low-density lipoprotein (LDL) cholesterol levels.

Table 1.

Characteristics of the study population according to tertiles of LA volume index (n = 672).

	Worse diastolic function	Intermediate diastolic function	Better diastolic function	p-linear
Demographics
Women	120 (53.6)	115 (51.3)	102 (45.5)	0.089
Age, years	69.9 ± 7.9	67.8 ± 6.6	67.4 ± 6.0	<0.001
Glucose metabolism status, NGM/IGM/T2DM	71/50/103 (40.8/25.0/34.2)	100/49/75 (44.6/21.9/33.5)	103/69/52 (46.0/30.8/23.2)	<0.001
Measures of (central) obesity
BMI, kg/m²	27.3 ± 3.7	27.5 ± 3.9	27.2 ± 3.5	0.843
Waist circumference, cm	95.5 ± 11.6	95.0 ± 11.0	95.0 ± 11.0	0.622
Waist-to-hip ratio	0.93 ± 0.11	0.92 ± 0.09	0.93 ± 0.09	0.321
Blood pressure
Systolic blood pressure, mmHg	143.9 ± 22.0	140.5 ± 19.7	140.3 ± 19.2	0.058
Diastolic blood pressure, mmHg	82.8 ± 11.3	83.5 ± 11.0	83.2 ± 10.5	0.717
Anti-hypertensive medication	116 (51.8)	74 (33.0)	59 (26.3)	<0.001
ACE inhibitors	34 (15.2)	19 (8.5)	18 (8.0)	0.014
Hypertension	170 (75.9)	149 (66.5)	142 (63.4)	0.004
Lipids
Total cholesterol, mmol/L	5.6 ± 1.0	5.7 ± 1.0	5.8 ± 1.1	0.021
High-density lipoprotein, mmol/L	1.4 ± 0.4	1.4 ± 0.4	1.4 ± 0.4	0.137
Total cholesterol/HDL ratio	4.2 ± 1.3	4.3 ± 1.2	4.5 ± 1.4	0.010
Low-density lipoprotein, mmol/L	3.5 ± 0.9	3.7 ± 0.9	3.7 ± 1.0	0.027
Triglycerides, mmol/L	1.3 [1.0–1.8]	1.3 [1.0–1.8]	1.4 [1.0–2.0]	0.009
Lipid-modifying medication	35 (15.6)	39 (17.4)	38 (17.0)	0.704
Glucose metabolism
HbA1c, %	6.0 ± 0.7	6.0 ± 0.8	6.0 ± 0.6	0.952
Fasting plasma glucose, mmol/L	6.5 ± 1.4	6.3 ± 1.5	6.2 ± 1.4	0.062
Oral anti-diabetics and/or insulin	15 (6.7)	14 (6.3)	8 (3.6)	0.147
Prior cardiovascular disease	124 (55.4)	102 (45.5)	92 (41.1)	0.002
Current smoking	27 (12.1)	31 (13.8)	39 (17.4)	0.107
Kidney function
Albuminuria	38 (17.0)	15 (6.7)	16 (7.1)	<0.001
eGFR (mL/min/1.73 m²)	64.6 ± 11.3	65.7 ± 9.6	64.2 ± 10.7	0.695
Advanced glycation endproducts
CML (nmol/mmol lysine)	33.6 [28.5–38.7]	32.1 [27.9–36.4]	31.6 [27.4–37.3]	0.062
CEL (nmol/mmol lysine)	34.7 [27.6–43.2]	34.4 [28.5–42.0]	33.9 [27.9–41.1]	0.621
Pentosidine (nmol/mmol lysine)	0.60 [0.47–0.77]	0.55 [0.45–0.65]	0.53 [0.45–0.65]	0.002

NGM: normal glucose metabolism; IGM: impaired glucose metabolism; T2DM: type 2 diabetes mellitus; BMI: body mass index; ACE: angiotensin-converting enzyme; HDL: high-density lipoprotein; HbA1c: haemoglobin A1c; eGFR: estimated glomerular filtration rate; CML: Nε-(carboxymethyl)lysine; CEL: Nε-(carboxyethyl)lysine; IQR: interquartile range; SD: standard deviation.

Data are presented as means ± SD, medians [IQR] or frequencies (percentages).

Table 2.

Echocardiographic indices of the study population according to tertiles of LA volume index (n = 672).

	Worse DF	Intermediate DF	Better DF	p-linear
Diastolic function
LA volume index, mL/m²	30.4 [27.3–35.2]	22.5 [21.4–23.8]	17.6 [16.1–18.8]	–
LA volume × LV mass index mL × g/m²	4382 [4308–7453]	3601 [2950–4382]	2652 [2202–3169]	–
Systolic function^a
LV ejection fraction, %	59.7 ± 9.8	62.5 ± 7.0	62.7 ± 7.2	<0.001
Left ventricular dimensions^b
LV end-diastolic diameter, mm	52.2 ± 6.5	50.3 ± 5.6	49.6 ± 5.2	<0.001
Inter ventricular septum, mm	10.3 ± 2.8	9.7 ± 2.2	9.4 ± 2.1	<0.001
Posterior wall thickness, mm	9.3 ± 1.6	8.9 ± 1.5	8.8 ± 1.4	<0.001
Relative wall thickness	0.36 ± 0.08	0.36 ± 0.08	0.36 ± 0.07	0.590
LV mass index, g/m²	102.9 ± 31.8	89.1 ± 22.9	84.0 ± 20.2	<0.001

LA: left atrial; LV: left ventricular; DF: diastolic function; SD: standard deviation; IQR: inter quartile range.

Data are presented as means ± SD or medians [IQR].

Missing per tertile 8/8/12.

Excluded individuals (n = 21).

With deteriorating diastolic function, serum pentosidine increased (p for linear trend 0.002), whereas no linear trend was observed with serum CML and CEL (p = 0.062 and p = 0.621, respectively) (Table 1). These associations were not materially altered after additional adjustment for age and/or waist circumference²⁵ (data not shown). Deteriorating diastolic function was associated with lower systolic function, greater LV end-diastolic diameter, increased wall thicknesses and a greater LVMI (p for linear trend all <0.001) (Table 2).

Associations of AGEs with diastolic function

After adjustment for the variables of model 2, higher levels of the Z-score of serum CML were, in NGM, associated with worse diastolic function. The regression coefficients [beta (95% confidence interval (CI))] for LA volume index and the LA volume × LVMI product term were −0.053 (−0.095; −0.011) and −0.064 (−0.126; −0.002), respectively. In IGM and T2DM, no significant associations were observed with both indices of diastolic function. The p-values for interaction between GMS and the Z-score of CML were 0.011 and 0.035 for the LA volume index and the LA volume × LVMI product term, respectively.

After adjustment for the variables of model 2, higher levels of the Z-score of serum pentosidine were, in NGM, associated with worse diastolic function in both indices (i.e. LA volume index and the LA volume × LVMI product term): −0.056 (−0.099; −0.013) and −0.084 (−0.417; −0.021), respectively. In IGM, lower levels of the Z-score of pentosidine were associated with lower levels of the LA volume × LVMI product term only: −0.080 (−0.164; 0.003). In T2DM, no significant associations could be observed between the Z-score of pentosidine and diastolic function. The p-values for interaction between GMS and Z-score of pentosidine were 0.040 and 0.085 for the LA volume index and the LA volume × LVMI product term, respectively.

Neither in NGM nor in IGM or in T2DM was the Z-score of serum CEL associated with diastolic function. The p-values for interaction between GMS and Z-score of CEL were 0.320 and 0.390 for the LA volume index and the LA volume × LVMI product term, respectively. (Table 3, Supplemental Figure 1(a)).

Table 3.

Associations of Z-score of serum AGEs with diastolic function (n = 672) and systolic function (n = 649) according to glucose metabolism status.

		M	CML		Pentosidine		CEL
		M	B (95% CI)	p-value	B (95% CI)	p-value	B (95% CI)	p-value
LN LA volume index, mL^a	NGM	1	−0.056 (−0.097;−0.014)	0.009	−0.058 (−0.101;−0.014)	0.009	0.004 (−0.034;0.043)	0.832
	NGM	2	−0.053 (−0.095;−0.011)	0.014	−0.056 (−0.099;−0.013)	0.011	0.004 (−0.034;0.043)	0.827
	IGM	1	−0.011 (−0.067;0.045)	0.699	−0.039 (−0.096;0.018)	0.177	−0.030 (−0.076;0.017)	0.211
	IGM	2	−0.005 (−0.061;0.052)	0.868	−0.033 (−0.091;0.024)	0.252	−0.026 (−0.072;0.020)	0.270
	T2DM	1	0.017 (−0.017;0.050)	0.323	−0.002 (−0.035;0.030)	0.899	0.029 (−0.010;0.067)	0.145
	T2DM	2	0.015 (−0.019;0.050)	0.380	−0.000 (−0.033;0.033)	0.987	0.015 (−0.024;0.054)	0.445
LN LA volume × LV mass index, mL × g/m^2,a	NGM	1	−0.037 (−0.100;0.027)	0.257	−0.073 (−0.138;−0.007)	0.030	−0.015 (−0.074;0.043)	0.608
	NGM	2	−0.064 (−0.126;−0.002)	0.043	−0.084 (−0.147;−0.021)	0.009	−0.007 (−0.064;0.049)	0.794
	IGM	1	−0.045 (−0.129;0.040)	0.302	−0.077 (−0.164;0.010)	0.082	−0.069 (−0.140;0.001)	0.055
	IGM	2	−0.061 (−0.144;0.022)	0.149	−0.080 (−0.164;0.003)	0.060	−0.042 (−0.110;0.025)	0.220
	T2DM	1	0.047 (−0.004;0.098)	0.071	−0.009 (−0.058;0.041)	0.726	0.047 (−0.012;0.106)	0.117
	T2DM	2	0.020 (−0.031;0.070)	0.444	−0.015 (−0.063;0.033)	0.533	0.027 (−0.030;0.084)	0.347
LV ejection fraction, %	NGM	1	−1.105 (−2.182;−0.027)	0.045	−0.500 (−1.628;0.628)	0.385	−0.968 (−1.977;0.040)	0.060
	NGM	2	−1.227 (−2.322;−0.131)	0.028	−0.494 (−1.626;0.637)	0.391	−0.971 (−1.976;0.034)	0.058
	IGM	1	0.042 (−1.448;1.533)	0.955	−0.243 (−1.802;1.316)	0.760	−0.228 (−1.479;1.022)	0.720
	IGM	2	−0.186 (−1.704;1.333)	0.810	−0.365 (−1.930;1.201)	0.648	−0.040 (−1.285;1.205)	0.950
	T2DM	1	1.167 (0.276;2.058)	0.010	0.253 (−0.617;1.124)	0.568	0.313 (−0.727;1.354)	0.555
	T2DM	2	0.950 (0.039;1.861)	0.041	0.230 (−0.650;1.110)	0.608	0.168 (−0.877;1.213)	0.753

NGM: normal glucose metabolism; IGM: impaired glucose metabolism; T2DM: type 2 diabetes mellitus; LA: left atrial; LV: left ventricular; LVMI: left ventricular mass index; CI: confidence interval; LN: natural logarithm; CML: Nε-(carboxymethyl)lysine; CEL: Nε-(carboxyethyl)lysine; HDL: high-density lipoprotein.

Model 1: adjusted for age, sex.

Model 2: model 1 + prior cardiovascular disease, hypertension and/or anti-hypertensive medication, triglycerides, total/HDL ratio, lipid-modifying medication, current smoking, estimated glomerular filtration rate and waist.

LA volume index and LA volume × LVMI product term values were reversed, that is, multiplied by −1 (lower values indicating worse diastolic function).

Associations of AGEs with systolic function

After adjustment for the variables of model 2, higher levels of the Z-score of serum CML were, in NGM, associated with lower LV ejection fraction [−1.227 (−2.322; −0.131)]. In IGM, no significant association could be observed; whereas in T2DM, higher levels of the Z-score of CML were associated with higher LV ejection fraction [0.950 (0.039; 1.861)]. The p-value for interaction between GMS and Z-score of CML for LV ejection fraction was 0.002. Neither in NGM nor in IGM or in T2DM was the Z-score of serum pentosidine associated with LV ejection fraction. The p-value for interaction between GMS and Z-score of pentosidine for LV ejection fraction was 0.316.

Higher levels of Z-score of serum CEL were associated with lower LV ejection fraction [−0.971 (−1.976; 0.034)] in NGM only. The p-value for interaction between GMS and Z-score of CEL for LV ejection fraction was 0.218 (Table 3 and Supplemental Figure 1(b)).

Associations of AGEs with LV structure

None of the serum Z-scores of the individual AGEs were associated with any of the individual indices of LV structure (Supplemental Table 1). No significant interaction between GMS and individual Z-scores of AGEs for the individual indices of LV structure were observed (data not shown).

Additional analyses

When the statistical analyses were repeated on those with full echocardiographic data (n = 644), the results did not materially change (data not shown). If we replaced waist circumference by either waist-to-hip ratio or body mass index, the results did not materially change (data not shown). Additional adjustments for insulin resistance, insulin therapy, RAAS-modifying agents or ACE inhibitors specifically did not materially change the results (data not shown).

Discussion

This study is the first large population-based cohort study, stratified according to GMS, that investigated the association between the protein-bound serum AGEs, CML, pentosidine and CEL on the one hand and LV function and structure as determined by echocardiography on the other.

The results showed that, in NGM only, higher levels of CML and pentosidine were associated with worse diastolic function. In IGM, higher levels of pentosidine were borderline and significantly associated with worse diastolic function; whereas in T2DM, none of the serum AGEs were associated with diastolic function. With regard to systolic function, in NGM, higher levels of CML and CEL were associated with worse systolic function. In IGM, no associations were observed; whereas in T2DM, higher levels of CML were associated with better systolic function. Finally, none of the serum AGEs were individually related to any of the individual indices of LV structure.

The results of this study extend previous knowledge by adding population-based data and by showing that serum AGEs are indeed associated with both impaired diastolic and systolic LV function in individuals with NGM, and thus may contribute to the pathobiology of heart failure. Such associations were, however, not observed, quite unexpectedly, in individuals with IGM or T2DM.

Interestingly, similar observations were made by Campbell et al.⁷ in patients undergoing coronary bypass surgery and by Berg et al.⁸ in individuals with type 1 diabetes. In the former study, the association between higher levels of circulating CML and diastolic dysfunction became stronger after exclusion of individuals with diabetes;⁷ whereas in the latter, higher levels of circulating CML were associated with better systolic function in individuals with diabetes.⁸

The above suggests that the associations between circulating AGEs and LV function and structure are not straightforward. Studies conducted on human^7,26–32 or animal^33–38 cardiac tissue have shown that AGEs accumulate within the myocardium,^7,26–34 even more so in T2DM^26–31,34 or IGM.³³ In addition, these studies showed that the myocardial accumulation of AGEs was associated with impaired diastolic function^30,33–36 and/or impaired systolic function.^36–38 The contrasting results of our study, where we measured circulating AGEs is the observation that they were not associated with LV function with deteriorating GMS and thereby calls into question whether circulating AGEs in IGM and T2DM may be representative of the accumulation of AGEs within the myocardium.

For example, it has been shown that, with increasing levels of obesity, CML becomes increasingly trapped in visceral adipose tissue, which was accompanied by lower levels of circulating CML and greater insulin resistance.²⁵ In the present study, however, adjustment for measures of obesity did not affect the results. Alternatively, it has been suggested that collagen, due to cross-linkage by AGEs, is less vulnerable to enzymatic degradation, and thereby traps AGEs in the extracellular matrix.³⁹ In addition, we cannot fully exclude the possibility that survival bias might have influenced our results to such an extent that the reported associations between circulating AGEs and LV function are underestimated, in particular, in individuals with IGM or T2DM, because of their increased cardiovascular mortality risk. In sum, a straightforward pathobiological explanation for the results of the present study and the studies by Campbell et al.⁷ and Berg et al.⁸ with respect to IGM and T2DM is not readily at hand, and this issue requires further study.

Nevertheless and regardless of how circulating AGEs may be (pathobiologically) handled with deteriorating GMS, an important role for AGEs in the pathobiology of heart failure can still be deducted from the fact that in ageing individuals with NGM, the risk of heart failure parallels the accumulation of AGEs.^2,40,41

Other limitations of this study may be its cross-sectional design, which does not allow strong causal inferences. In addition, the criteria to define diastolic function have changed considerably over the years. Nevertheless, LA volume index and the LA volume × LVMI product term have been shown to be reliable proxy measurements for chronic diastolic function.^19,20

In conclusion, in this population-based cohort study, in NGM, both higher levels of serum CML and pentosidine were associated with worse diastolic function, and higher levels of CML and CEL were associated with worse systolic function. In IGM, a similar pattern emerged, though less consistent; whereas in T2DM, these associations were either non-existent or even reversed. The results of this study thereby suggests that serum AGEs are associated with impaired LV function in individuals with NGM but that in individuals with deteriorating GMS serum AGEs may not mirror heart failure risk.

Footnotes

Declaration of conflicting interests

P. Linssen, C. Schalkwijk, J. Dekker, G. Nijpels and C. Stehouwer have nothing to disclose. R. Henry reports grants and personal fees from Sanofi, outside the submitted work. HP. Brunner-La Rocca reports grants and personal fees from Roche Diagnostics, grants from AstraZeneca, grants from GSK and personal fees from Novartis, outside the submitted work.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

References

Kannel

McGee

. Diabetes and cardiovascular disease. The Framingham study. JAMA 1979; 241: 2035–2038.

Hartog

Voors

Bakker

. Advanced glycation end-products (AGEs) and heart failure: pathophysiology and clinical implications. Eur J Heart Fail 2007; 9: 1146–1155.

Bodiga

Eda

Bodiga

. Advanced glycation end products: role in pathology of diabetic cardiomyopathy. Heart Fail Rev 2014; 19: 49–63.

Borbely

Papp

Edes

. Molecular determinants of heart failure with normal left ventricular ejection fraction. Pharmacol Rep 2009; 61: 139–145.

Brownlee

. Biochemistry and molecular cell biology of diabetic complications. Nature 2001; 414: 813–820.

Daffu

del Pozo

O’Shea

. Radical roles for RAGE in the pathogenesis of oxidative stress in cardiovascular diseases and beyond. Int J Mol Sci 2013; 14: 19891–19910.

Campbell

Somaratne

Jenkins

. Diastolic dysfunction of aging is independent of myocardial structure but associated with plasma advanced glycation end-product levels. PLoS ONE 2012; 7: e49813.

Berg

Snorgaard

Faber

. Serum levels of advanced glycation end products are associated with left ventricular diastolic function in patients with type 1 diabetes. Diabetes Care 1999; 22: 1186–1190.

Sveen

Nerdrum

Hanssen

. Impaired left ventricular function and myocardial blood flow reserve in patients with long-term type 1 diabetes and no significant coronary artery disease: associations with protein glycation. Diab Vasc Dis Res 2014; 11: 84–91.

10.

Hartog

Hummel

Voors

. Skin-autofluorescence, a measure of tissue advanced glycation end-products (AGEs), is related to diastolic function in dialysis patients. J Card Fail 2008; 14: 596–602.

11.

Steine

Larsen

Stugaard

. LV systolic impairment in patients with asymptomatic coronary heart disease and type 1 diabetes is related to coronary atherosclerosis, glycaemic control and advanced glycation endproducts. Eur J Heart Fail 2007; 9: 1044–1050.

12.

Mooy

Grootenhuis

de Vries

. Prevalence and determinants of glucose intolerance in a Dutch caucasian population. The Hoorn Study. Diabetes Care 1995; 18: 1270–1273.

13.

Spijkerman

Adriaanse

Dekker

. Diabetic patients detected by population-based stepwise screening already have a diabetic cardiovascular risk profile. Diabetes Care 2002; 25: 1784–1789.

14.

Henry

Kostense

Spijkerman

. Arterial stiffness increases with deteriorating glucose tolerance status: the Hoorn Study. Circulation 2003; 107: 2089–2095.

15.

Hanssen

Engelen

Ferreira

. Plasma levels of advanced glycation endproducts Nε-(carboxymethyl)lysine, Nε-(carboxyethyl)lysine, and pentosidine are not independently associated with cardiovascular disease in individuals with or without type 2 diabetes: the Hoorn and CODAM studies. J Clin Endocrinol Metab 2013; 98: E1369–E1373.

16.

Scheijen

van de Waarenburg

Stehouwer

. Measurement of pentosidine in human plasma protein by a single-column high-performance liquid chromatography method with fluorescence detection. J Chromatogr B Analyt Technol Biomed Life Sci 2009; 877: 610–614.

17.

Henry

Kamp

Kostense

. Left ventricular mass increases with deteriorating glucose tolerance, especially in women: independence of increased arterial stiffness or decreased flow-mediated dilation: the Hoorn study. Diabetes Care 2004; 27: 522–529.

18.

Lang

Bierig

Devereux

. Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005; 18: 1440–1463.

19.

Pritchett

Mahoney

Jacobsen

. Diastolic dysfunction and left atrial volume: a population-based study. J Am Coll Cardiol 2005; 45: 87–92.

20.

Melenovsky

Borlaug

Rosen

. Cardiovascular features of heart failure with preserved ejection fraction versus nonfailing hypertensive left ventricular hypertrophy in the urban Baltimore community: the role of atrial remodeling/dysfunction. J Am Coll Cardiol 2007; 49: 198–207.

21.

Alberti

Zimmet

. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 1998; 15: 539–553.

22.

1999 World Health Organization-International Society of Hypertension Guidelines for the Management of Hypertension. Guidelines Subcommittee. J Hypertens 1999; 17: 151–183.

23.

Levey

Bosch

Lewis

. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999; 130: 461–470.

24.

Matthews

Hosker

Rudenski

. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419.

25.

Gaens

Goossens

Niessen

. Nε-(carboxymethyl)lysine-receptor for advanced glycation end product axis is a key modulator of obesity-induced dysregulation of adipokine expression and insulin resistance. Arterioscler Thromb Vasc Biol 2014; 34: 1199–1208.

26.

Schalkwijk

Baidoshvili

Stehouwer

. Increased accumulation of the glycoxidation product Nepsilon-(carboxymethyl)lysine in hearts of diabetic patients: generation and characterisation of a monoclonal anti-CML antibody. Biochim Biophys Acta 2004; 1636: 82–89.

27.

Donaldson

Taatjes

Zile

. Combined immunoelectron microscopic and computer-assisted image analyses to detect advanced glycation end-products in human myocardium. Histochem Cell Biol 2010; 134: 23–30.

28.

Nozynski

Zakliczynski

Konecka-Mrowka

. Advanced glycation end-products in myocardium-supported vessels: effects of heart failure and diabetes mellitus. J Heart Lung Transplant 2011; 30: 558–564.

29.

Nozynski

Zakliczynski

Konecka-Mrowka

. Advanced glycation end product accumulation in the cardiomyocytes of heart failure patients with and without diabetes. Ann Transplant 2012; 17: 53–61.

30.

Van Heerebeek

Hamdani

Handoko

. Diastolic stiffness of the failing diabetic heart: importance of fibrosis, advanced glycation end products, and myocyte resting tension. Circulation 2008; 117: 43–51.

31.

Falcao-Pires

Hamdani

Borbely

. Diabetes mellitus worsens diastolic left ventricular dysfunction in aortic stenosis through altered myocardial structure and cardiomyocyte stiffness. Circulation 2011; 124: 1151–1159.

32.

Campbell

Somaratne

Jenkins

. Impact of type 2 diabetes and the metabolic syndrome on myocardial structure and microvasculature of men with coronary artery disease. Cardiovasc Diabetol 2011; 10: 80.

33.

Avendano

Agarwal

Bashey

. Effects of glucose intolerance on myocardial function and collagen-linked glycation. Diabetes 1999; 48: 1443–1447.

34.

Schafer

Huber

Wihler

. Impaired left ventricular relaxation in type 2 diabetic rats is related to myocardial accumulation of N(epsilon)-(carboxymethyl) lysine. Eur J Heart Fail 2006; 8: 2–6.

35.

Norton

Candy

Woodiwiss

. Aminoguanidine prevents the decreased myocardial compliance produced by streptozotocin-induced diabetes mellitus in rats. Circulation 1996; 93: 1905–1912.

36.

Falcao-Pires

Palladini

Goncalves

. Distinct mechanisms for diastolic dysfunction in diabetes mellitus and chronic pressure-overload. Basic Res Cardiol 2011; 106: 801–814.

37.

Petrova

Yamamoto

Muraki

. Advanced glycation endproduct-induced calcium handling impairment in mouse cardiac myocytes. J Mol Cell Cardiol 2002; 34: 1425–1431.

38.

Yan

Luo

. Effects of advanced glycation end products on calcium handling in cardiomyocytes. Cardiology 2014; 129: 75–83.

39.

Nowotny

Grune

. Degradation of oxidized and glycoxidized collagen: role of collagen cross-linking. Arch Biochem Biophys 2014; 542: 56–64.

40.

Lakatta

. Arterial and cardiac aging: major shareholders in cardiovascular disease enterprises: part III: cellular and molecular clues to heart and arterial aging. Circulation 2003; 107: 490–497.

41.

Lakatta

Levy

. Arterial and cardiac aging: major shareholders in cardiovascular disease enterprises: part II: the aging heart in health: links to heart disease. Circulation 2003; 107: 346–354.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.24 MB