Influencing factors on cardiac structure and function beyond glycemic control in patients with type 2 diabetes mellitus

A total of 148 consecutive asymptomatic patients with T2DM and without cardiovascular disease, who were admitted to our institution (Juntendo University Hospital Tokyo, Japan) for diabetic educational program between January 2010 and May 2012, were prospectively enrolled in this study. Patients were included if they met the following inclusion criteria: no symptoms and history of heart disease, LV ejection fraction (LVEF) > 50%, absence of regional LV wall motion abnormalities, and clinically stable. Exclusion criteria were coronary artery disease, congenital heart disease, atrial fibrillation, significant heart valve disease, renal failure (serum creatinine > 2.0 mg/dl) and type 1 diabetes mellitus. In addition, 68 age-matched healthy controls without hypertension, dyslipidemia or diabetes mellitus served as a control group. The study protocol was approved by the Institutional Review Board of Juntendo University Hospital.

This study was a prospective cohort study. The diagnoses of hypertension, dyslipidemia, coronary artery disease and cerebrovascular accident were assessed by the treating physician. Standard comprehensive two-dimensional and Doppler echocardiographic examinations were performed using commercially available systems. As indicators of obesity, body mass index was calculated from height and weight and the waist-hip ratio was calculated from the waist and hip circumference.

In all patients and control subjects, cardiac chamber quantification by 2D echocardiography was performed according to guidelines provided by the American Society of Echocardiography [25]. LV diameters were measured using 2D echocardiography according to the recommended criteria. The thickness of the intraventricular septum and the LV posterior wall were measured at end-diastole. LV mass was calculated using diastolic measurements of LV diameter and wall thickness on 2D echocardiography according to the formula recommended by the American Society of Echocardiography [25]. LV end-diastolic volume and end-systolic volume were determined from the apical views using modified biplane Simpson’s method. LV ejection fraction was calculated by the following equation: 100 × (end-diastolic volume - end-systolic volume)/end-diastolic volume. Each pmeter was indexed for body surface area (BSA), when appropriate.

For assessing conventional diastolic pmeters, left atrial volume was measured by Simpson’s biplane method (indexed to body surface area), and mitral inflow velocities were determined by pulsed Doppler and tissue Doppler imaging. The peak early (E) and late (A) diastolic velocity, the deceleration time from the peak of the early diastolic wave to baseline (DCT) and the E/A ratio were assessed from the mitral inflow velocity pattern. The mitral annular motion velocity was recorded at the medial mitral annulus site in the apical 4-chamber view by pulsed tissue Doppler echocardiography. Peak early (e’) and late (a’) diastolic velocity of the annulus were measured and the ratio of peak early diastolic transmitral flow velocity to annular velocity (E/e’) was calculated [26]. Furthermore, according to previous reports [26, 27], we classified diastolic function in all subjects into one of four categories (normal, grade I, grade II or grade III) to compare the prevalence of diastolic dysfunction between healthy subjects and diabetic patients.

Blood samples for determination of serum creatinine, fasting blood sugar, glycosylated hemoglobin, immunoreactive insulin, C-peptide immunoreactivity, N-terminal pro B-type natriuretic peptide (NT-pro BNP) and lipid profiles (triglycerides, high-density lipoprotein cholesterol and low-density lipoprotein cholesterol) were drawn in patients with T2DM. Moreover, the homeostasis model assessment ratio (HOMA-R) index was calculated as fasting plasma glucose × fasting plasma insulin/22.5 for assessing insulin resistance [28]. Non-esterified fatty acid (NEFA) was measured since it has been suggested to play a critical role in triggering the development of cellular insulin resistance and myocardial contractile dysfunction [11]. In addition, high-sensitive C-reactive protein (hsCRP) was measured as a marker of inflammation. Estimated glomerular filtration rate (eGFR) was determined based on the new Japanese coefficient-modified Modification of Diet in Renal Disease (MDRD) study equation [29]. The formula is as follows: eGFR=194 × serum creatinine (SCr)^-1.094 × age-0.287, where age is in years, SCr is in mg/dL and GFR is in mL/min/1.73 m² body surface area. The product of this equation was multiplied by a correction factor of 0.739 in women.

VAT and subcutaneous adipose tissue (SAT) were measured quantitatively by computed tomography (CT) (Aquilion64, Toshiba, Tokyo, Japan), as previously reported [30]. Briefly, an axial CT scan at the level of the umbilicus was obtained for each participant using an electron beam CT scanner (commercial available software). Planimetric measurements at the level of the umbilicus have been shown to correlate well with volumetric quantification of VAT (r=0.81 in men and r=0.85 in women, p<0.001) and SAT (r=0.95 in men and r=0.85 in women, p<0.001) [30]. The images generated were transferred to a workstation and analyzed using commercial software. The coefficients of variation between two observers analyzing the same VAT and SAT images (n=30) ranged from 0.6% to 14.2% and from 0.1% to 7.3%, respectively. Both VAT and SAT was indexed for body surface area as well as echocardiographic pmeters.

All statistical analyses were performed using SPSS version 17.0 (SPSS, Inc., Chicago, IL). Data are expressed as mean ± standard deviation (SD). The relationships between variables were assessed using Pearson’s correlation coefficient. We compared baseline characteristics and echo pmeters between the control group and patients with T2DM using an unpaired t-test. Categorical data were compared between the two groups using a chi-square test. When we found echocardiographic pmeters that were different between the diabetics and controls, Pearson’s linear correlation analysis was used to determine the correlations between these echocardiographic pmeters and the clinical data in just the patients with T2DM. Multiple linear regression analysis was also performed using the clinical data to evaluate the independent determinants of these echocardiographic pmeters in the T2DM patients. A log-transformation was used to normalize the distributions of triglyceride, HOMA-R and NT-proBNP level. A P-value <.05 was considered statistically significant.

Table 2 shows differences in echocardiography pmeters between the two groups. Patients with T2DM had a greater LV mass index than normal controls. LV mass index of control subjects, diabetic patients without hypertension, and diabetic patients with hypertension, were 70±12, 79±18, and 83±22 g/m², respectively. LV mass index of both diabetic patients with and without hypertension were increased as compared with control subjects (p<0.01 and p<0.001, respectively) as shown in previous studies [5‐8]. Patients with T2DM also had a significantly larger left atrial volume index, lower e’ and E/A, and higher E/e’, indicating LV diastolic dysfunction, as previously reported [5‐8]. Whereas 23% of the age-matched healthy controls had diastolic dysfunction with advancing age, 56% of patients with T2DM had diastolic dysfunction, indicating a higher prevalence of diastolic dysfunction in the diabetic population (Figure 1). None of the subjects were classified into grade III diastolic dysfunction.

Laboratory data and visceral fat area measured by CT scanning are shown in Table 3. Among the clinical factors in Tables 1 and 3, we determined which factors were associated with the abnormal echocardiographic pmeters shown in Table 2. The LV mass index had a significant correlation with systolic blood pressure, waist/hip ratio, low-density cholesterol level and VAT/SAT ratio (Table 4). Multivariate liner regression analysis found that systolic blood pressure, waist/hip ratio and VAT/SAT ratio were independent factors associated with LV mass index (Table 5).

The correlation between clinical factors and the diastolic pmeters are also shown in Table 4. These diastolic pmeters were mainly correlated with gender, age, hypertension and VAT, whereas E/A and e’ were also correlated with decreased eGFR. In multivariate linear regression analysis, age was the strongest independent determinant of left atrial volume, E/A, e’ and E/e’ among all the clinical variables. Furthermore, VAT was also an independent determinant of these four diastolic pmeters (Table 5, Figure 2), indicating excess VAT may be independently associated with diastolic dysfunction in T2DM. As for e’ and E/e’, female gender and hypertension were independent determinants. Furthermore, there were positive correlations between VAT and log triglyceride (r=0.376, p<0.001), IRI (r=0.393, p<0.001) and log HOMA-R (r=0.363, p<0.001).

While we included asymptomatic patients without overt cardiovascular disease, we did not completely exclude latent patients with coronary artery disease using stress testing or coronary angiography. We did not measure serum adipocytokine levels and could not evaluate adipose dysfunction. Further investigation is required to determine the direct association between VAT and the level of adipocytokine dysfunction in diabetic cardiomyopathy. We included patients with the comorbidity of hypertension in this study. Hypertension caused by activation of the renin-angiotensin system in diabetes mellitus plays an important role in the development of cardiac dysfunction in T2DM. In addition, patients with T2DM have high prevalence of hypertension in the real clinical setting and are more subject to LV hypertrophy caused by hypertension than those without diabetes [43]. These characteristics are important aspects of cardiac dysfunction in T2DM. Instead, we performed multiple regression analysis using clinical factors including history of hypertension and blood pressure to evaluate the independent determinants of LV mass and diastolic dysfunction, and VAT or VAT/SAT ratio was an independent determinant of cardiac structure and function in our population after adjustment for history of hypertension and blood pressure. However, hypertension per se causes diastolic dysfunction and LV hypertrophy, and further investigation in normotensive patients with T2DM is desirable to more accurately assess the mechanisms of cardiac abnormalities in T2DM. Moreover, further investigation in comparison with patients with hypertension but without T2DM would provide new insights to this issue. Finally, we measured only LV ejection fraction as an indicator of systolic function in this study. Ernande et al [17]. reported reduced systolic strain using speckle-tracking imaging in T2DM; therefore, further investigation of systolic dysfunction with speckle-tracking imaging would be of interest.