Abstract
Background
The diabetic heart exhibits increased left ventricular (LV) mass and reduced ventricular function. However, this relationship has not been studied in patients with aortic stenosis (AS), a disease process that causes LV hypertrophy and dysfunction through a distinct mechanism of pressure overload. The aim of this study was to determine how diabetes mellitus (DM) impacts LV remodeling and function in patients with severe AS.
Methods and Results
Echocardiograms were performed on 114 patients with severe AS [mean aortic valve area (AVA) 0.6 cm2] and included measures of LV remodeling and function. Multivariable linear regression models investigated the independent effect of DM on these aspects of LV structure and function. Compared to non-diabetics (n=60), diabetics (n=54) had increased LV mass, LV end-systolic dimension, LV end-diastolic dimension, and decreased LV ejection fraction (EF) and longitudinal systolic strain (p<0.01 for all). In multivariable analyses adjusting for age, sex, systolic BP, AVA, BSA, and coronary disease, DM was an independent predictor of increased LV mass (β=26g, p=0.01), LV end-systolic dimension (β=0.5cm, p=0.008), and LV end-diastolic dimension (β=0.3cm, p=0.025). After additionally adjusting for LV mass, DM was associated with reduced longitudinal systolic strain (β=1.9%, p=0.023) and a trend toward reduced EF (β=−5%, p=0.09). Among diabetics, insulin use (as a marker of disease severity) was associated with larger LV end-systolic dimension and worse LV function. LV mass was a strong predictor of reduced EF and systolic strain (p<0.001 for both).
Conclusions
DM has an additive adverse effect on hypertrophic remodeling—increased LV mass and larger cavity dimensions—and is associated with reduced systolic function in patients with AS beyond known factors of pressure overload.
Keywords: aortic stenosis, diabetes mellitus, LV hypertrophy, LV remodeling, echocardiography
Diabetes mellitus (DM) increases the risk of developing, and exacerbates the complications of, all forms of cardiovascular disease.1,2 DM can directly affect the myocardium leading to increased left ventricular (LV) hypertrophy, altered LV remodeling, and reduced LV function.3 In the presence of hypertension or ischemia, DM exacerbates adverse LV remodeling and dysfunction, which results in worse heart failure symptoms and increased mortality.4–6 As the prevalence of DM reaches epidemic levels, the relationship between DM and LV remodeling and heart failure becomes increasingly important to understand.
In patients with AS, LV hypertrophy is associated with reduced systolic and diastolic function,7 increased operative mortality,8 and worse long-term outcomes.9 Prior studies of LV hypertrophic remodeling in patients with AS have predominantly investigated the impact of pressure overload—aortic valve area (AVA), valve gradients, blood pressure, arterial stiffness—and gender.10–12 Despite evidence that the diabetic heart exhibits increased LV mass and reduced ventricular function, most studies have focused on the impact of DM on the development and progression of valve calcification and stenosis.13,14 The role of DM in adverse LV remodeling has not been explored in patients with AS.
The importance of investigating the influence of DM on LV remodeling in patients with AS is supported by recent evidence that patients with AS and insulin resistance—in the form of metabolic syndrome—have an increased LV mass index and reduced LV function.15 In addition, preclinical data using a mouse model of diet-induced obesity demonstrated that animals fed a high-fat diet, which caused insulin resistance and hyperglycemia, developed more pronounced LV hypertrophy and dysfunction after aortic banding than mice fed a standard diet.16
Given the known effects of DM on the LV in non-AS patients and emerging data about the influence of metabolic disturbances on LV remodeling, we hypothesized that DM adversely impacts LV remodeling and LV function in patients with AS beyond traditional factors of pressure overload.
Methods
Patient population
Between February 2008 and June 2010, 114 consecutive study participants were recruited from patients referred for high-risk aortic valve replacement and scheduled for a clinically-indicated echocardiogram to evaluate severe symptomatic AS at Barnes Jewish Hospital in St. Louis, Missouri. Patients had to be ≥18 years of age and have an AVAindex < 0.6 cm2/m2. Patients with irregular rhythms or echocardiographic windows that precluded accurate two-dimensional (2D) measurements were excluded. Institutional Review Board approval was obtained, and all patients signed a written informed consent.
Clinical data
Clinical variables were obtained through patient interview and chart abstraction. The diagnosis of diabetes mellitus was determined from the medical record and confirmed by patient interview. Prior myocardial infarction was defined as pathologic Q waves on ECG, history of elevated troponin thought to be related to an acute coronary syndrome, history of myocardial infarction as stated in the medical record, or evidence of infarct on nuclear imaging. Coronary artery disease was defined as prior myocardial infarction (as above), prior percutaneous or surgical revascularization, or any stenosis in the LAD, LCX, RCA, or left main ≥ 50%. The diagnosis of hypertension was made either by medical history, if the patient was taking antihypertensive medications, or if the untreated blood pressure was >140/90. Blood pressure and NYHA functional class were determined at the time of the echocardiogram.
Echocardiographic data
Parasternal and apical views were used to acquire standard 2D and Doppler images on a GE Vivid 7 ultrasound system using a multifrequency transducer with tissue Doppler capability. Echocardiograms were analyzed by three cardiologists blinded to the patients’ diabetic status. The severity of AS was determined by measuring mean and peak gradients across the valve using the modified Bernoulli equation and by calculating AVA using the continuity equation. AVA was indexed to body surface area (BSA). Systemic arterial compliance and valvulo-arterial impedance were calculated as previously described.12 The severity of mitral regurgitation was determined by the integrative approach recommended by the American Society of Echocardiography.17
LV remodeling and function
The LV minor axis internal dimensions were measured at end-systole (LVESD) and end-diastole (LVEDD) in the parasternal long-axis view. Posterior wall thickness (PWT) was measured at end-diastole in the same view. LV mass was calculated by the area-length method and indexed to BSA. Relative wall thickness (RWT) = (2 × PWT)/LVEDD. Stroke volume = π (LVOT radius)2 × LVOT VTI and was indexed to BSA. LV ejection fraction (LV EF) was measured using the biplane Simpson’s method. The septal and lateral early diastolic mitral annular velocity (e′) was acquired using tissue Doppler imaging from the apical 4-chamber view; the average was reported. E/e’ represents the peak E wave velocity from the pulse Doppler of mitral valve inflow divided by the averaged e′. All measurements reflect an average of several values and were made in accordance with the recommendations of the American Society of Echocardiography.18
Longitudinal systolic strain and strain rate
Standard 2D apical 2-, 3-, and 4-chamber views obtained at a frame rate of 60–80 were analyzed with GE EchoPac analysis software (versions 7.2 and 108.1.5; GE Vingmed Ultrasound A.S., Horten, Norway) to measure strain and strain rate using the speckle tracking method. In each apical view, the LV endocardium was traced at end systole, and the region of interest width was adjusted to fit the myocardial wall thickness. The software automatically tracked the motion through the rest of the cardiac cycle. The beginning of systole was established as the end of the QRS by ECG analysis, whereas the timing of end systole was determined using continuous wave Doppler through the aortic valve. Adequate tracking of the ventricular myocardium was verified in real time, and inadequately tracked segments were excluded from analysis. LV peak longitudinal systolic strain and peak systolic strain rate were measured in all 18 segments of myocardium.19 Strain and strain rate were averaged across the 6 segments within the basal, mid, and apical regions of the LV. The values for the 3 regions were then averaged, yielding a global average for the entire myocardium. If more than 1 segment per region had inadequate tracking, the patient was excluded from strain analysis.
Statistical analysis
Demographic, clinical, and echocardiographic characteristics were compared between diabetics and non-diabetics using chi-square test for categorical variables and t-test for continuous variables. NYHA Class was compared between groups using the Cochran-Armitage trend test.
To assess the effect of DM on LV remodeling, multivariable linear regression models were constructed to evaluate the impact of DM on LV mass (primary structural variable), LV end-systolic dimension (LVESD), LV end-diastolic dimension (LVEDD), and relative wall thickness (RWT). Balancing increased discrimination with over-fitting, covariates were selected a priori based on clinical judgment and included DM, age, sex, BSA, systolic blood pressure, AVA, and coronary disease. BSA was selected as a covariate given its relationship with indexing LV mass and AVA. However, given the strong univariable and known physiologic association between BMI and LV mass, we performed sensitivity analyses that included BMI as a covariate instead of BSA (both were not included given collinearity).
To assess the effect of DM on LV function, multivariable linear regression models were constructed to evaluate the impact of DM on longitudinal systolic strain (primary functional variable), LV EF, average e’, and longitudinal strain rate. Covariates in functional models included the same variables as the structural models with the addition of LV mass, the primary structural variable.
As an exploratory analysis, an ordinal logistic regression model was constructed to evaluate the effect of DM on heart failure symptoms. Due to small numbers, patients with NYHA Class I and II symptoms were combined into a single group. Covariates in the model included the same variables as the functional models with the addition of systolic strain as the primary functional variable. Ordinal logistic regression, which allows the outcome variable to have >2 categories, assumes a proportional odds ratio (OR) for each predictor for each combination of higher-risk categories versus lower-risk categories (e.g. NYHA Class IV vs. NYHA Class I-III and NYHA Class III-IV vs. NYHA Class I-II). The validity of the proportional odds assumption, assessed with the Score test,20 was met for the model.
Due to issues with multiple comparisons, a primary structural outcome (LV mass) and primary functional outcome (LV systolic strain) were selected and a Bonferroni correction was used. Accordingly, a p-value of <0.025 was considered to be statistically significance for these analyses. All other tests of statistical significance were evaluated at a 2-sided significance level of 0.05 with 95% confidence intervals (CIs). All statistical analyses were performed using SAS for Windows version 9.2 (SAS Institute, Inc., Cary, NC).
Results
The primary goal of our study was to determine the association of DM with LV remodeling and LV function in patients with severe AS.
Patient population
Among 114 patients with severe symptomatic AS included in this study, the average age was 82 years, 54 (47%) were female, mean AVA was 0.6 cm2, mean EF was 50%, and 54 (47%) were diabetic. The baseline demographic and clinical characteristics of those with and without DM are shown in Table 1. Diabetics were younger, more often male, had larger body mass indices, and higher B-type natriuretic peptide levels than non-diabetics. The prevalence of coronary disease, prior infarct, and hypertension were similar in diabetics and non-diabetics, as were the levels of blood pressure, renal function, and cholesterol. As expected, diabetic patients were more frequently prescribed angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers. The severity of AS by valve area and gradients was similar between diabetics and non-diabetics (Table 2).
Table 1.
Clinical Characteristics
| Diabetics | Non-Diabetics | p-value | |
|---|---|---|---|
| n=54 | n=60 | ||
| Age (years) | 79.9 (6.7) | 83.9 (7.8) | 0.004 |
| Female sex | 35.2% | 58.3% | 0.014 |
| Caucasian race | 100% | 100% | 1.0 |
| Prior myocardial infarction* | 33.3% | 31.7% | 0.85 |
| Prior coronary artery disease† | 72.2% | 65.0% | 0.41 |
| Hypertension | 87.0% | 87.7% | 0.95 |
| Systolic blood pressure (mmHg) | 132.5 (27.0) | 135.1 (23.5) | 0.59 |
| Diastolic blood pressure (mmHg) | 68.9 (10.3) | 69.0 (10.5) | 0.96 |
| Body mass index (kg/m2) | 30.0 (7.7) | 26.3 (5.1) | 0.003 |
| Body surface area (m2) | 2.00 (0.27) | 1.83 (0.28) | <0.001 |
| NYHA Class | 0.16 | ||
| I-II | 5.8% | 14.6% | |
| III | 65.4% | 63.6% | |
| IV | 28.9% | 21.8% | |
| Glomerular filtration rate (mL/min/1.73m2)‡ | 55.5 (24.1) | 60.3 (19.6) | 0.24 |
| B-type natriuretic peptide (pg/mL) | 907.1 (931.1) | 513.4 (491.4) | 0.009 |
| Total cholesterol (mg/dL) | 131.1 (43.2) | 133.6 (45.5) | 0.79 |
| Triglycerides (mg/dL) | 106.7 (70.5) | 100.1 (49.5) | 0.61 |
| High density lipoprotein (mg/dL) | 40.1 (15.4) | 43.0 (14.5) | 0.35 |
| Low density lipoprotein (mg/dL) | 69.9 (10.3) | 72.2 (35.3) | 0.75 |
| Hemoglobin A1c (%)§ | 7.1 (1.2) | NA | NA |
| Diabetic treatment | NA | ||
| Diet-controlled | 22.2% | NA | |
| Oral medications only | 33.3% | NA | |
| Insulin | 44.4% | NA | |
| ACE-inhibitors or ARBs | 64.8% | 51.7% | 0.041 |
| Beta blockers | 70.4% | 68.3% | 0.81 |
| Statins | 77.8% | 60.0% | 0.16 |
All values are mean (SD) unless otherwise specified
Clinical history of myocardial infarction, elevated troponin thought to be related to an acute coronary syndrome, infarct on nuclear imaging, or pathologic Q waves on ECG
Myocardial infarction, percutaneous or surgical revascularization, or any stenosis in the LAD, LCX, RCA, or Left Main ≥ 50%
Estimated by the Modification of Diet in Renal Disease Study method
Hemoglobin A1c data was obtained on 65% of the diabetic patients
Table 2.
Echocardiographic Parameters
| Diabetics | Non-Diabetics | p-value | |
|---|---|---|---|
| n=54 | n=60 | ||
| Aortic Stenosis Severity | |||
| Aortic valve area (cm2) | 0.61 (0.16) | 0.60 (0.17) | 0.72 |
| Aortic valve area index (cm2/m2) | 0.30 (0.07) | 0.32 (0.07) | 0.085 |
| Mean gradient (mmHg) | 38.5 (11.7) | 39.6 (12.0) | 0.60 |
| Peak gradient (mmHg) | 62.7 (18.3) | 64.6 (19.0) | 0.59 |
| Left Ventricular Structural Parameters | |||
| LV end-systolic dimension (cm) | 3.73 (0.96) | 3.01 (1.18) | <0.001 |
| LV end-diastolic dimension (cm) | 4.95 (0.63) | 4.40 (0.90) | <0.001 |
| Posterior wall thickness (cm) | 1.30 (0.18) | 1.23 (0.18) | 0.034 |
| Relative wall thickness | 0.54 (0.11) | 0.58 (0.14) | 0.069 |
| LV mass (g) | 269.0 (50.4) | 222.3 (63.4) | <0.001 |
| LV mass index (g/m2) | 135.6 (24.8) | 121.4 (26.1) | 0.004 |
| Left Ventricular Functional Parameters | |||
| Stroke volume index (mL/m2) | 28.5 (7.1) | 31.0 (7.4) | 0.075 |
| LV ejection fraction (%) | 45.4 (17.1) | 54.5 (17.5) | 0.006 |
| Average e’ (cm/s) | 4.84 (1.59) | 4.89 (1.42) | 0.88 |
| E/e’ | 27.7 (12.9) | 23.0 (14.3) | 0.079 |
| Systolic strain (%) | −12.0 (4.8) | −15.1 (5.0) | 0.001 |
| Strain rate (s−1) | −0.83 (0.24) | −0.94 (0.26) | 0.024 |
| Other Parameters | |||
| Systemic arterial compliance index* | 0.49 (0.18) | 0.50 (0.15) | 0.84 |
| Valvulo-arterial impedance† | 6.28 (1.61) | 5.88 (1.34) | 0.15 |
| Mitral regurgitation severity‡ | 2.02 (0.71) | 1.92 (0.74) | 0.46 |
All values are mean (SD)
Stroke volume index/pulse pressure
(Systolic blood pressure + mean gradient)/stroke volume index
Graded on a 5-point scale (0=none to 4=severe)
LV Remodeling
Table 2 shows that there were several differences between diabetics and non-diabetics with regard to LV remodeling. Compared to non-diabetics, diabetic patients exhibited increased LV mass (DM vs. non-DM: 269g vs. 222g, p<0.001), larger LVESD (DM vs. non-DM: 3.7cm vs. 3.0cm, p<0.001), larger LVEDD (DM vs. non-DM: 5.0cm vs. 4.4cm, p<0.001), and increased PWT (DM vs. non-DM: 1.3cm vs. 1.2cm, p=0.034). Because the increase in chamber dimensions was more marked than the increase in PWT among diabetics, the RWT trended towards being larger among non-diabetics.
Multivariable analyses were performed to evaluate the independent effect of DM on LV remodeling beyond traditional factors of pressure overload (Table 3). After controlling for age, gender, BSA, systolic blood pressure, AVA, and coronary disease, the left ventricles of diabetics were, on average, 26 grams heavier (95% CI 6.6 to 45.0, p=0.01), 0.5 cm larger in diameter at end-systole (95% CI 0.15 to 0.93, p=0.008), and 0.3 cm larger in diameter at end-diastole (95% CI 0.04 to 0.60, p=0.025) than the left ventricles of non-diabetics. DM was not associated with RWT after multivariable adjustment (p=0.51). When these analyses were performed substituting BMI for BSA—to ensure that our results were not confounded by the effect of obesity on LV remodeling—the relationship between DM and adverse LV remodeling was slightly stronger.
Table 3.
Effect of Diabetes Mellitus on Left Ventricular Remodeling
| Independent Variables | LV Mass | LVESD | LVEDD | RWT | ||||
|---|---|---|---|---|---|---|---|---|
| β-estimate (g) | p-value | β-estimate(cm) | p-value | β-estimate (cm) | p-value | β-estimate (unitless) | p-value | |
| Diabetes mellitus | 25.8 | 0.010 | 0.54 | 0.008 | 0.32 | 0.025 | −0.016 | 0.51 |
| Age | 0.1 | 0.89 | −0.01 | 0.52 | −0.02 | 0.16 | 0.003 | 0.11 |
| Female sex | −11.6 | 0.37 | −0.43 | 0.12 | −0.37 | 0.052 | 0.072 | 0.028 |
| Body surface area | 91.2 | <0.001 | −0.05 | 0.93 | 0.38 | 0.28 | 0.004 | 0.95 |
| Systolic blood pressure (per 10 mmHg) | −0.2 | 0.92 | −0.12 | 0.002 | −0.06 | 0.015 | 0.015 | 0.001 |
| Aortic valve area | 32.0 | 0.38 | 0.75 | 0.31 | 0.39 | 0.45 | −0.003 | 0.98 |
| Coronary disease | 17.2 | 0.13 | 0.28 | 0.23 | 0.11 | 0.51 | −0.007 | 0.80 |
Abbreviations: LVESD=left ventricular end-systolic dimension, LVEDD=left ventricular end-diastolic dimension; RWT=relative wall thickness
The impact of insulin use (as a marker of disease severity) on LV remodeling was assessed in the diabetic patients. In univariable analysis, diabetics on insulin vs. no insulin trended toward larger LV mass (283g vs. 259g, p=0.09), although this association was not significant after multivariable adjustment (p=0.47). Diabetics on insulin vs. no insulin had larger LVESD (4.1cm vs. 3.5cm, p=0.02); after adjusting for factors known to affect LV remodeling, insulin use was associated with an increased LVESD (β=0.6cm, p=0.02). There was no association, either in univariable or multivariable analyses, of insulin use with LVEDD or RWT.
LV Function
Compared to non-diabetics, patients with DM had evidence of decreased systolic function, with worse longitudinal systolic strain (DM vs. non-DM: −12% vs. −15%, p=0.001), LV EF (DM vs. non-DM: 45% vs. 55%, p=0.006), and longitudinal systolic strain rate (DM vs. non-DM: −0.8s−1 vs. −0.9s−1, p=0.024) (Table 2). Diastolic function as measured by e’ was similar between the groups, but there was a trend towards an increased filling pressure in diabetics as measured by E/e’ (DM vs. non-DM: 28 vs. 23, p=0.079). Measures of global LV load and vascular compliance were similar between the groups, as was the severity of mitral regurgitation.
To evaluate the independent effect of DM on LV function beyond factors of pressure overload and LV remodeling, multivariable analyses were performed to control for variables known to influence LV function (Table 4). After adjusting for the same covariates as above in addition to LV mass, longitudinal systolic strain was, on average, 1.9% worse (95% CI 0.3 to 3.6, p=0.023), and EF was 5.3% worse in diabetics than in non-diabetics, although the association between DM and EF did not reach statistical significance (95% CI -11.4 to 0.8, p=0.09). Longitudinal systolic strain rate (p = 0.11) and e’ (p=0.75) were not associated with DM after multivariable adjustment. LV mass was strongly associated with all measures of reduced systolic and diastolic function (p≤0.01 for all analyses; Table 4). Again, when BMI was substituted for BSA in the multivariable analyses of LV function, the relationship between DM and reduced LV function was slightly stronger.
Table 4.
Effect of Diabetes Mellitus on Left Ventricular Function
| Independent Variables | Systolic Strain | LV Ejection Fraction | Average e’ | Strain Rate | ||||
|---|---|---|---|---|---|---|---|---|
| β-estimate (%) | p-value | β-estimate (%) | p-value | β-estimate (cm/s) | p-value | β-estimate (s−1) | p-value | |
| Diabetes mellitus | 1.93 | 0.023 | −5.3 | 0.093 | 0.10 | 0.75 | 0.07 | 0.11 |
| Age | −0.04 | 0.52 | 0.02 | 0.93 | −0.01 | 0.66 | 0.001 | 0.71 |
| Female sex | 0.28 | 0.80 | 5.2 | 0.19 | −0.74 | 0.061 | −0.06 | 0.31 |
| Body surface area | −4.58 | 0.042 | 22.0 | 0.006 | 1.26 | 0.10 | −0.45 | <0.001 |
| Systolic blood pressure (per 10 mmHg) | −0.79 | <0.001 | 2.3 | <0.001 | 0.12 | 0.037 | −0.03 | <0.001 |
| Aortic valve area | −5.74 | 0.066 | 3.3 | 0.77 | 0.80 | 0.47 | −0.13 | 0.42 |
| Coronary disease | 1.79 | 0.056 | −1.5 | 0.67 | −0.53 | 0.13 | 0.10 | 0.037 |
| LV mass (per 10 g) | 0.42 | <0.001 | −1.4 | <0.001 | −0.10 | 0.002 | 0.02 | <0.001 |
The impact of insulin use on LV function was also assessed among diabetics. Diabetics on insulin vs. no insulin had worse systolic strain (−10.0% vs. −13.5%, p=0.01), lower LVEF (40.0% vs. 49.7%, p=0.04), and worse strain rate (−0.74s−1 vs. −0.89s−1, p=0.03). After multivariable adjustment, insulin use was associated with worse systolic strain (β=2.4%, p=0.05), lower LVEF (β=−10.9%, p=0.01), and a trend toward worse strain rate (p=0.06). There was no association of insulin use with average e′.
NYHA Class
Given the influence of DM on LV structure and function, an exploratory analysis was performed to determine the independent effect of DM on functional capacity in patients with severe AS. After controlling for the same covariates as above in addition to LV systolic strain, there was a trend toward diabetics having a worse NYHA class than non-diabetics (OR 2.5, 95% CI 0.97–6.67, p=0.059; Table 5).
Table 5.
Effect of Diabetes Mellitus on Heart Failure Symptoms
| Independent Variables | Worse NYHA Class | |
|---|---|---|
| OR* (95% CI) | p-value | |
| Diabetes mellitus | 2.54 (0.97–6.67) | 0.059 |
| Age | 1.08 (0.99–1.16) | 0.14 |
| Female sex | 4.43 (1.26–15.51) | 0.020 |
| Body surface area | 6.05 (0.47–77.86) | 0.17 |
| Systolic blood pressure (per 10mmHg) | 0.98 (0.81–1.19) | 0.86 |
| Aortic valve area | 0.14 (0.004–4.75) | 0.27 |
| Coronary disease | 3.52 (1.21–10.26) | 0.021 |
| LV mass (per 10g) | 1.00 (0.90–1.11) | 0.98 |
| Systolic strain | 1.03 (0.92–1.15) | 0.62 |
Odds ratio is estimating the odds of being in one higher NYHA class using ordinal logistic regression (NYHA Class I-II vs. Class III vs. Class IV)
Discussion
In patients with severe AS, DM has an additive adverse effect on LV remodeling and LV function, with increased hypertrophic remodeling—increased LV mass and larger cavity dimensions—and reduced LV systolic function. These effects remained significant even after controlling for factors of pressure overload and gender that are known to influence LV remodeling and function in patients with AS. Furthermore, we found that increased LV mass was a strong predictor of decreased LV systolic and diastolic function, which is consistent with multiple prior studies that have demonstrated that pronounced LV hypertrophic remodeling in patients with severe AS is maladaptive.7–9,21 Since DM had a deleterious effect on LV function independent of increased LV mass and body size, our study suggests that DM adversely impacts LV function through a combination of increased LV mass and additional undefined deleterious effects on the myocardium itself. Finally, we found that diabetics also trended toward having decreased functional capacity that appeared to be independent of an effect of LV structure and function. To our knowledge, this is the first report on the impact of DM on LV structure and function in patients with AS.
Diabetes and the Left Ventricle – Comparison with Prior Studies
Prior studies in patients without AS have demonstrated that DM is associated with increased LV mass, higher absolute and relative wall thickness, and decreased LV systolic and diastolic function.3,22 Diabetics also exhibit relatively greater increases in LV hypertrophy and dysfunction in the face of systemic hypertension and ischemia/infarction than non-diabetics; these changes are associated with worse heart failure symptoms and outcomes in diabetic patients.4–6 Our results are consistent with these prior observations and extend these findings to patients with both DM and AS, a population that has not previously been well characterized.
Our study also expands on the recent report published by Page et al.,15 which evaluated the effects of metabolic syndrome on LV structure and function in patients with asymptomatic mild-to-moderate AS. Both investigations demonstrated that metabolic abnormalities are associated with increased LV mass index and reduced LV function in patients with AS. However, they noted that metabolic syndrome was associated with a different manner of remodeling in patients with less severe AS: increased absolute and RWT but similar chamber dimensions. The differences in LV remodeling seen by Page et al. may be due to the type/severity of metabolic disturbance (DM vs. metabolic syndrome), the severity of AS (severe symptomatic AS vs. mild-to-moderate asymptomatic AS), or other factors. Considering our results in combination with their study, however, raises the possibility that in the face of a metabolic disturbance (DM, metabolic syndrome, or both), the LV initially responds to the pressure overload with an increase in LV mass and RWT, but over time as the AS progresses, the LV of diabetics is likely to decompensate more rapidly, with increased LV dilation and reduced LV function. Further longitudinal studies are needed to characterize the influence of DM and metabolic syndrome (alone and together) on LV remodeling and function as AS progresses.
Potential Mechanisms for Adverse Remodeling and Decreased Function Among Diabetics with AS
Although this study was not designed to determine the molecular mechanisms by which DM affects the myocardium, there are several potential mechanistic explanations for our findings. Type 2 DM is characterized by a milieu of hyperinsulinemia, insulin resistance, hyperglycemia, and increased nonesterified fatty acids, which contribute to lipotoxicity, oxidative stress, advanced glycation end products (AGEs), and altered calcium handling and substrate metabolism.23–26 In the myocardium, these changes promote fibrosis, apoptosis, and myocyte hypertrophy, which have structural and functional consequences.24 Pressure overload also induces changes in molecular signaling pathways that lead to similar deleterious structural and functional changes.27 If and how these two stimuli—DM and AS—interact to influence signaling pathways and myocardial structure and function is not known; this is an area ripe for further evaluation, particularly given the findings of our study.
Clinical Implications
Given the well-documented adverse effects of LV hypertrophy in patients with AS, our findings of the additive effect of DM on LV remodeling and LV function have important clinical implications. Previous reports have focused on the impact of the stenotic valve on this remodeling process, which currently is only modifiable through surgical intervention with valve replacement. Our study has identified a risk factor for accentuated LV hypertrophy in patients with AS that may be a therapeutic target. Given the adverse impact of DM in patients with AS, future studies should evaluate whether better DM control—perhaps with specific medications—prevents adverse remodeling and its attendant consequences.
Additionally, because adverse remodeling and reduced LV function are known to impact outcomes, our findings that diabetics are more likely to have increased LV mass and LV dilation with decreased systolic function may influence decisions on the timing and type of valve replacement. In patients with severe AS, DM has been associated with higher mortality.28 At our own institution, an analysis of 1080 patients with AS over the last decade showed that DM was an independent predictor of decreased survival after aortic valve replacement (Ralph J. Damiano, Jr., MD, unpublished data, 2010). It is tempting to speculate that earlier surgical intervention in diabetics with severe AS (perhaps while they are asymptomatic) may reduce this disparity in outcomes; however, this would require prospective evaluation in a lower risk group of subjects. Furthermore, there may be implications for the type of procedure chosen for valve implantation—open surgery vs. transcatheter—as DM may influence how the LV responds to either cardioplegia vs. rapid ventricular pacing, particularly in patients with underlying LV dysfunction.
DM may also impact the post-valve replacement clinical course. After aortic valve replacement, regression of LV hypertrophy is associated with greater improvement in LV contractility.29 As diabetics with hypertension exhibit less regression of LV hypertrophy than non-diabetics when treated with anti-hypertensive medications,30 diabetics may also experience less regression of LV hypertrophy or less improvement of LV function after valve replacement, which would be expected to impact symptomatic recovery. All of these questions require further exploration with answers that will impact our care of diabetic patients with AS.
Limitations
Our study should be interpreted in the context of the following limitations. This is a single-center experience at a tertiary care facility, which could cause referral or other biases and limit the generalizability of the findings. Second, this study is a cross-sectional analysis of patients with severe AS, which limits our ability to know how DM affects the LV response to AS over time as the AS worsens. Future longitudinal studies will be needed to address this question and clarify whether DM has an additive or synergistic effect beyond AS in the LV remodeling process. Third, we did not have data on fasting blood glucose or waist circumference, which prevented us from classifying subjects according to the presence/absence of metabolic syndrome. Moreover, beyond the need for insulin, we were not able to characterize the severity of DM in terms of length of time with the disease or associated microvascular complications. We were able to obtain HbA1c data on 65% of the diabetic patients, but this only reflects three months of DM control and does not reflect the long-term exposure of the LV to the metabolic effects of DM. Of note, the mean HbA1c in our study was 7.1, suggesting that even in this relatively well-controlled diabetic population (at least at the time of their echocardiograms), the differences in LV remodeling and function were still evident. To extend our initial observation of this relationship, it will be important to obtain data on the severity of DM as well as insulin, glucose, and nonesterified fatty acid levels for future studies seeking to clarify and characterize the relationship between DM and LV remodeling in patients with AS. Finally, given the observational nature of the study, there is the potential for unmeasured confounding. Nevertheless, the findings are consistent with the existing literature on DM and LV remodeling.
Conclusion
This study shows for the first time that DM is associated with increased LV hypertrophy, increased chamber dimensions, and reduced systolic function in patients with severe AS, even after controlling for known factors of pressure overload, body size, and other potential confounders. These data extend previous observations about the deleterious influence of DM on LV remodeling and function to a population of patients with AS, a disease process that results in LV hypertrophy and dysfunction through pressure overload. Our observations should stimulate further studies that explore underlying mechanisms, implications for clinical management, and opportunities for pharmacologic intervention. A better understanding of how DM influences the LV in patients with AS will be critical for improving management of these high-risk patients.
Acknowledgments
The authors thank Michelle Myers and Christina MacNaughton for their help with data collection and Joel D. Schilling, MD, PhD, and Lisa de las Fuentes, MD, MS, for reviewing the manuscript.
Funding Sources: This study was supported by a grant from the American Heart Association Midwest Affiliate Clinical Research Program (09CRP210070) to Dr. Lindman and by NIH / National Center for Research Resources (NCRR) Washington University-ICTS Grants (KL2 RR024994 and UL1 RR024992). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official view of NCRR or NIH. Dr. Arnold is supported by the Outcomes Research Postdoctoral Fellowship awarded by the American Heart Association Pharmaceutical Roundtable and David and Stevie Spina.
Footnotes
Disclosures: None
References
- 1.Kannel WB, McGee DL. Diabetes and cardiovascular disease. The Framingham study. JAMA. 1979;241:2035–2038. doi: 10.1001/jama.241.19.2035. [DOI] [PubMed] [Google Scholar]
- 2.Mak KH, Topol EJ. Emerging concepts in the management of acute myocardial infarction in patients with diabetes mellitus. J Am Coll Cardiol. 2000;35:563–568. doi: 10.1016/s0735-1097(99)00628-2. [DOI] [PubMed] [Google Scholar]
- 3.Devereux RB, Roman MJ, Paranicas M, O’Grady MJ, Lee ET, Welty TK, Fabsitz RR, Robbins D, Rhoades ER, Howard BV. Impact of diabetes on cardiac structure and function: the strong heart study. Circulation. 2000;101:2271–2276. doi: 10.1161/01.cir.101.19.2271. [DOI] [PubMed] [Google Scholar]
- 4.Abbott RD, Donahue RP, Kannel WB, Wilson PW. The impact of diabetes on survival following myocardial infarction in men vs women. The Framingham Study. JAMA. 1988;260:3456–3460. [PubMed] [Google Scholar]
- 5.Bella JN, Devereux RB, Roman MJ, Palmieri V, Liu JE, Paranicas M, Welty TK, Lee ET, Fabsitz RR, Howard BV. Separate and joint effects of systemic hypertension and diabetes mellitus on left ventricular structure and function in American Indians (the Strong Heart Study) Am J Cardiol. 2001;87:1260–1265. doi: 10.1016/s0002-9149(01)01516-8. [DOI] [PubMed] [Google Scholar]
- 6.Palmieri V, Bella JN, Arnett DK, Liu JE, Oberman A, Schuck MY, Kitzman DW, Hopkins PN, Morgan D, Rao DC, Devereux RB. Effect of type 2 diabetes mellitus on left ventricular geometry and systolic function in hypertensive subjects: Hypertension Genetic Epidemiology Network (HyperGEN) study. Circulation. 2001;103:102–107. doi: 10.1161/01.cir.103.1.102. [DOI] [PubMed] [Google Scholar]
- 7.Kupari M, Turto H, Lommi J. Left ventricular hypertrophy in aortic valve stenosis: preventive or promotive of systolic dysfunction and heart failure? Eur Heart J. 2005;26:1790–1796. doi: 10.1093/eurheartj/ehi290. [DOI] [PubMed] [Google Scholar]
- 8.Fuster RG, Argudo JA, Albarova OG, Sos FH, Lopez SC, Sorli MJ, Codoner MB, Minano JA. Left ventricular mass index in aortic valve surgery: a new index for early valve replacement? Eur J Cardiothorac Surg. 2003;23:696–702. doi: 10.1016/s1010-7940(03)00092-7. [DOI] [PubMed] [Google Scholar]
- 9.Mihaljevic T, Nowicki ER, Rajeswaran J, Blackstone EH, Lagazzi L, Thomas J, Lytle BW, Cosgrove DM. Survival after valve replacement for aortic stenosis: implications for decision making. J Thorac Cardiovasc Surg. 2008;135:1270–1278. doi: 10.1016/j.jtcvs.2007.12.042. discussion 1278–1279. [DOI] [PubMed] [Google Scholar]
- 10.Cramariuc D, Rieck AE, Staal EM, Wachtell K, Eriksen E, Rossebo AB, Gerdts E. Factors influencing left ventricular structure and stress-corrected systolic function in men and women with asymptomatic aortic valve stenosis (a SEAS Substudy) Am J Cardiol. 2008;101:510–515. doi: 10.1016/j.amjcard.2007.09.100. [DOI] [PubMed] [Google Scholar]
- 11.Bech-Hanssen O, Wallentin I, Houltz E, Beckman Suurkula M, Larsson S, Caidahl K. Gender differences in patients with severe aortic stenosis: impact on preoperative left ventricular geometry and function, as well as early postoperative morbidity and mortality. Eur J Cardiothorac Surg. 1999;15:24–30. doi: 10.1016/s1010-7940(98)00268-1. [DOI] [PubMed] [Google Scholar]
- 12.Briand M, Dumesnil JG, Kadem L, Tongue AG, Rieu R, Garcia D, Pibarot P. Reduced systemic arterial compliance impacts significantly on left ventricular afterload and function in aortic stenosis: implications for diagnosis and treatment. J Am Coll Cardiol. 2005;46:291–298. doi: 10.1016/j.jacc.2004.10.081. [DOI] [PubMed] [Google Scholar]
- 13.Katz R, Wong ND, Kronmal R, Takasu J, Shavelle DM, Probstfield JL, Bertoni AG, Budoff MJ, O’Brien KD. Features of the metabolic syndrome and diabetes mellitus as predictors of aortic valve calcification in the Multi-Ethnic Study of Atherosclerosis. Circulation. 2006;113:2113–2119. doi: 10.1161/CIRCULATIONAHA.105.598086. [DOI] [PubMed] [Google Scholar]
- 14.Kamalesh M, Ng C, El Masry H, Eckert G, Sawada S. Does diabetes accelerate progression of calcific aortic stenosis? Eur J Echocardiogr. 2009;10:723–725. doi: 10.1093/ejechocard/jep048. [DOI] [PubMed] [Google Scholar]
- 15.Page A, Dumesnil JG, Clavel MA, Chan KL, Teo KK, Tam JW, Mathieu P, Despres JP, Pibarot P. Metabolic syndrome is associated with more pronounced impairment of left ventricle geometry and function in patients with calcific aortic stenosis: a substudy of the ASTRONOMER (Aortic Stenosis Progression Observation Measuring Effects of Rosuvastatin) J Am Coll Cardiol. 55:1867–1874. doi: 10.1016/j.jacc.2009.11.083. [DOI] [PubMed] [Google Scholar]
- 16.Raher MJ, Thibault HB, Buys ES, Kuruppu D, Shimizu N, Brownell AL, Blake SL, Rieusset J, Kaneki M, Derumeaux G, Picard MH, Bloch KD, Scherrer-Crosbie M. A short duration of high-fat diet induces insulin resistance and predisposes to adverse left ventricular remodeling after pressure overload. Am J Physiol Heart Circ Physiol. 2008;295:H2495–2502. doi: 10.1152/ajpheart.00139.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD, Levine RA, Nihoyannopoulos P, Otto CM, Quinones MA, Rakowski H, Stewart WJ, Waggoner A, Weissman NJ. Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr. 2003;16:777–802. doi: 10.1016/S0894-7317(03)00335-3. [DOI] [PubMed] [Google Scholar]
- 18.Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, Picard MH, Roman MJ, Seward J, Shanewise JS, Solomon SD, Spencer KT, Sutton MS, Stewart WJ. 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. doi: 10.1016/j.echo.2005.10.005. [DOI] [PubMed] [Google Scholar]
- 19.Reisner SA, Lysyansky P, Agmon Y, Mutlak D, Lessick J, Friedman Z. Global longitudinal strain: a novel index of left ventricular systolic function. J Am Soc Echocardiogr. 2004;17:630–633. doi: 10.1016/j.echo.2004.02.011. [DOI] [PubMed] [Google Scholar]
- 20.Allison PD. Logistic Regression Using the SAS System: Theory and Application. Cary, NC: SAS Publishing; 1999. [Google Scholar]
- 21.Frey N, Katus HA, Olson EN, Hill JA. Hypertrophy of the heart: a new therapeutic target? Circulation. 2004;109:1580–1589. doi: 10.1161/01.CIR.0000120390.68287.BB. [DOI] [PubMed] [Google Scholar]
- 22.Galderisi M, Anderson KM, Wilson PW, Levy D. Echocardiographic evidence for the existence of a distinct diabetic cardiomyopathy (the Framingham Heart Study) Am J Cardiol. 1991;68:85–89. doi: 10.1016/0002-9149(91)90716-x. [DOI] [PubMed] [Google Scholar]
- 23.Cai L, Wang Y, Zhou G, Chen T, Song Y, Li X, Kang YJ. Attenuation by metallothionein of early cardiac cell death via suppression of mitochondrial oxidative stress results in a prevention of diabetic cardiomyopathy. J Am Coll Cardiol. 2006;48:1688–1697. doi: 10.1016/j.jacc.2006.07.022. [DOI] [PubMed] [Google Scholar]
- 24.Poornima IG, Parikh P, Shannon RP. Diabetic cardiomyopathy: the search for a unifying hypothesis. Circ Res. 2006;98:596–605. doi: 10.1161/01.RES.0000207406.94146.c2. [DOI] [PubMed] [Google Scholar]
- 25.Stanley WC, Lopaschuk GD, McCormack JG. Regulation of energy substrate metabolism in the diabetic heart. Cardiovasc Res. 1997;34:25–33. doi: 10.1016/s0008-6363(97)00047-3. [DOI] [PubMed] [Google Scholar]
- 26.Petrova R, Yamamoto Y, Muraki K, Yonekura H, Sakurai S, Watanabe T, Li H, Takeuchi M, Makita Z, Kato I, Takasawa S, Okamoto H, Imaizumi Y, Yamamoto H. Advanced glycation endproduct-induced calcium handling impairment in mouse cardiac myocytes. J Mol Cell Cardiol. 2002;34:1425–1431. doi: 10.1006/jmcc.2002.2084. [DOI] [PubMed] [Google Scholar]
- 27.Takimoto E, Champion HC, Li M, Ren S, Rodriguez ER, Tavazzi B, Lazzarino G, Paolocci N, Gabrielson KL, Wang Y, Kass DA. Oxidant stress from nitric oxide synthase-3 uncoupling stimulates cardiac pathologic remodeling from chronic pressure load. J Clin Invest. 2005;115:1221–1231. doi: 10.1172/JCI21968. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Brown ML, Pellikka PA, Schaff HV, Scott CG, Mullany CJ, Sundt TM, Dearani JA, Daly RC, Orszulak TA. The benefits of early valve replacement in asymptomatic patients with severe aortic stenosis. J Thorac Cardiovasc Surg. 2008;135:308–315. doi: 10.1016/j.jtcvs.2007.08.058. [DOI] [PubMed] [Google Scholar]
- 29.Taniguchi K, Takahashi T, Toda K, Matsue H, Shudo Y, Shintani H, Mitsuno M, Sawa Y. Left ventricular mass: impact on left ventricular contractile function and its reversibility in patients undergoing aortic valve replacement. Eur J Cardiothorac Surg. 2007;32:588–595. doi: 10.1016/j.ejcts.2007.07.003. [DOI] [PubMed] [Google Scholar]
- 30.Gerdts E, Okin PM, Omvik P, Wachtell K, Dahlof B, Hildebrandt P, Nieminen MS, Devereux RB. Impact of diabetes on treatment-induced changes in left ventricular structure and function in hypertensive patients with left ventricular hypertrophy. The LIFE study. Nutr Metab Cardiovasc Dis. 2009;19:306–312. doi: 10.1016/j.numecd.2008.12.009. [DOI] [PubMed] [Google Scholar]