Diastolic dysfunction of aging is independent of myocardial structure but associated with plasma advanced glycation end-product levels

Abstract

Background: Heart failure is associated with abnormalities of myocardial structure, and plasma levels of the advanced glycation end-product (AGE) N(ε)-(carboxymethyl)lysine (CML) correlate with the severity and prognosis of heart failure. Aging is associated with diastolic dysfunction and increased risk of heart failure, and we investigated the hypothesis that diastolic dysfunction of aging humans is associated with altered myocardial structure and plasma AGE levels.

Methods: We performed histological analysis of non-ischemic left ventricular myocardial biopsies and measured plasma levels of the AGEs CML and low molecular weight fluorophores (LMWFs) in 26 men undergoing coronary artery bypass graft surgery who had transthoracic echocardiography before surgery. None had previous cardiac surgery, myocardial infarction, atrial fibrillation, or heart failure.

Results: The patients were aged 43-78 years and increasing age was associated with echocardiographic indices of diastolic dysfunction, with higher mitral Doppler flow velocity A wave (r = 0.50, P = 0.02), lower mitral E/A wave ratio (r = 0.64, P = 0.001), longer mitral valve deceleration time (r = 0.42, P = 0.03) and lower early diastolic peak velocity of the mitral septal annulus, e' (r = 0.55, P = 0.008). However, neither mitral E/A ratio nor mitral septal e' was correlated with myocardial total, interstitial or perivascular fibrosis (picrosirius red), immunostaining for collagens I and III, CML, and receptor for AGEs (RAGE), cardiomyocyte width, capillary length density, diffusion radius or arteriolar dimensions. Plasma AGE levels were not associated with age. However, plasma CML levels were associated with E/A ratio (r = 0.44, P = 0.04) and e' (r = 0.51, P = 0.02) and LMWF levels were associated with E/A ratio (r = 0.49, P = 0.02). Moreover, the mitral E/A ratio remained correlated with plasma LMWF levels in all patients (P = 0.04) and the mitral septal e' remained correlated with plasma CML levels in non-diabetic patients (P = 0.007) when age was a covariate.

Conclusions: Diastolic dysfunction of aging was independent of myocardial structure but was associated with plasma AGE levels.

Figure 1. Correlations between echocardiographic parameters and age in men with coronary artery disease.

Age was not correlated with mitral Doppler flow velocity E wave velocity (A), but was correlated with mitral Doppler flow velocity A wave velocity (B), E/A ratio (C), mitral early inflow deceleration time (D) and mitral early diastolic peak velocity of the septal mitral annulus, e’ (E). However, the correlation of age with E/e’ ratio was not statistically significant (F).

Figure 2. Picrosirius-red staining of collagen, reticulin staining of cardiomyocyte membranes, and CD31 immunostaining of capillaries.

Representative sections of left ventricular biopsies from coronary artery bypass graft surgery patients stained with picrosirius-red demonstrating interstitial and perivascular fibrosis (stained red) and arteriolar dimensions (A), reticulin stain demonstrating cardiomyocyte membranes (B), and immunostained for CD31 demonstrating capillaries (C).

Figure 3. Correlations between myocardial fibrosis and echocardiographic parameters of diastolic dysfunction, mitral E/A ratio and mitral septal e

’ velocity, in men with coronary artery disease. Myocardial total fibrosis, interstitial fibrosis and perivascular fibrosis were not correlated with mitral E/A ratio (A, B, C) or with mitral septal e’ velocity (D, E, F).

Figure 4. Immunostaining for collagens I and III.

Representative sections of a left ventricular biopsy from a coronary artery bypass graft surgery patient immunostained for collagen I (A), collagen III (B), and a negative control section without primary antibody (C).

Figure 5. Correlations between collagens I and III as assessed by immunostaining and echocardiographic parameters of diastolic dysfunction, mitral E/A ratio and mitral septal e

’ velocity, in men with coronary artery disease. Total collagens I and III were not correlated with mitral E/A ratio (A, B) or with mitral septal e’ velocity (D, E) whereas the correlations between collagen I/collagen III ratio and mitral E/A ratio (C) and mitral septal e’ velocity (F) were of borderline statistical significance.

Figure 6. Correlations between cardiomyocyte width, capillary length density and arteriolar wall area/circumference ratio and echocardiographic parameters of diastolic dysfunction, mitral E/A ratio and mitral septal e

’ velocity, in men with coronary artery disease. Cardiomyocyte width, capillary length density and arteriolar wall area/circumference ratio were not correlated with mitral E/A ratio (A, B, C) or with mitral septal e’ velocity (D, E, F).

Figure 7. Immunostaining for N^ε-(carboxymethyl)lysine (CML) and the receptor for advanced glycation end-products (RAGE).

Representative sections of left ventricular biopsies from coronary artery bypass graft surgery patients immunostained for CML (A) or RAGE (C) and their corresponding negative control sections (B, D).

Figure 8. Correlations between N^ε-(carboxymethyl)lysine (CML) as assessed by immunostaining and echocardiographic parameters of diastolic dysfunction, mitral E/A ratio and mitral septal e

’ velocity, in men with coronary artery disease. Immunostaining for CML in arteriolar media and intima was not correlated with mitral E/A ratio (A, B) or with mitral septal e’ velocity (D, E).

Figure 9. Correlations between the receptor for advanced glycation end-products (RAGE) as assessed by immunostaining and echocardiographic parameters of diastolic dysfunction, mitral E/A ratio and mitral septal e

’ velocity, in men with coronary artery disease. Immunostaining for RAGE in arteriolar media and intima was not correlated with mitral E/A ratio (A, B) or with mitral septal e’ velocity (D, E).

Figure 10. Correlations between plasma advanced glycation end-products and echocardiographic parameters of diastolic dysfunction, mitral E/A ratio and mitral septal e

’ velocity, in men with coronary artery disease. Plasma N^ε-(carboxymethyl)lysine (CML) concentrations were correlated with mitral E/A ratio (A) and mitral septal e’ velocity (B), both for all patients and after exclusion of diabetic patients (non-diabetic patients). In addition, plasma low molecular weight fluorophore (LMWF) concentrations were correlated with mitral E/A ratio (C) for all patients and after exclusion of diabetic patients. Individual values are shown for non-diabetic (open circles) and diabetic patients (closed circles). Correlations are shown for all patients (dotted line) and for non-diabetic patients (dashed line).