Author + information
- Received April 26, 2002
- Revision received August 26, 2002
- Accepted November 1, 2002
- Published online February 19, 2003.
- Zhi You Fang, BM, MHSc*,
- Satoshi Yuda, MD, PhD*,
- Vinah Anderson, BSN*,
- Leanne Short, BS*,
- Colin Case, MSc* and
- Thomas H Marwick, MB.BS, PhD, FACC*,* ()
- ↵*Reprint requests and correspondence:
Prof. Thomas H. Marwick, University of Queensland Department of Medicine, Princess Alexandra Hospital, Ipswich Road, Brisbane, Qld 4012, Australia.
Objectives We sought to determine whether disturbances of myocardial contractility and reflectivity could be detected in diabetic patients without overt heart disease and whether these changes were independent and incremental to left ventricular hypertrophy (LVH).
Background Left ventricular (LV) dysfunction is associated with diabetes mellitus, but LVH is common in this population and the relationship between diabetic LV dysfunction and LVH is unclear.
Methods We studied 186 patients with normal ejection fraction and no evidence of CAD: 48 with diabetes mellitus only (DM group), 45 with LVH only (LVH group), 45 with both diabetes and LVH (DH group), and 48 normal controls. Peak strain and strain rate of six walls in apical four-chamber, long-axis, and two-chamber views were evaluated and averaged for each patient. Calibrated integrated backscatter (IB) was assessed by comparison of the septal or posterior wall with pericardial IB intensity.
Results All patient groups (DM, DH, LVH) showed reduced systolic function compared with controls, evidenced by lower peak strain (p < 0.001) and strain rate (p = 0.005). Calibrated IB, signifying myocardial reflectivity, was greater in each patient group than in controls (p < 0.05). Peak strain and strain rate were significantly lower in the DH group than in those in the DM alone (p < 0.03) or LVH alone (p = 0.01) groups.
Conclusions Diabetic patients without overt heart disease demonstrate evidence of systolic dysfunction and increased myocardial reflectivity. Although these changes are similar to those caused by LVH, they are independent and incremental to the effects of LVH.
Diabetes mellitus is associated with heart failure (1), largely through its association with hypertension and coronary artery disease (CAD). However, the presence of primary myocardial disease in diabetic patients has been anticipated from evidence of systolic and diastolic left ventricular (LV) dysfunction, even in the absence of other known cardiac disease (2–5). Interstitial fibrosis has been documented in biopsy studies of diabetic patients without hypertension or CAD (2,6), and changes due to fibrosis have been documented noninvasively with echocardiography. In less selected patients (without excluding contributions from hypertrophy and coronary disease), calibrated integrated backscatter (IB) (a marker of myocardial reflectivity), have been shown to be abnormal in the hearts of diabetic subjects (7,8). Myocardial strain and strain rate are new markers that may be used to measure regional myocardial function (9–12), but they have not been applied in diabetic subjects.
Left ventricular hypertrophy (LVH) and hypertension are common comorbidities of diabetes mellitus. Indeed, the combination of diabetes and hypertension is particularly damaging to the heart, and drug studies have shown a significant effect of small increments of blood pressure control in diabetic populations (13). We sought to determine whether the recorded changes in the diabetic heart were independent of LVH, after excluding a significant contribution from CAD. This study used sensitive markers of systolic performance and myocardial reflectivity to investigate early myocardial changes in patients with diabetes and normal systolic function in diabetic patients with and without LVH.
We studied 186 subjects (93 with and 93 without diabetes mellitus) who were eligible for recruitment if they had an ejection fraction >50% as assessed by the modified biplane Simpson’s method, no history of CAD, and a normal stress echocardiogram and/or coronary angiogram. We excluded patients with abnormal systolic function, moderate-to-severe valvular disease, CAD, atrial fibrillation or other severe arrhythmias, and congenital heart disease. The 186 subjects were divided into four groups: 48 with diabetes mellitus only (DM group; 33 male, 15 female, mean age 60 ± 10 years); 45 with both diabetes and LVH (DH group; 16 male, 29 female, mean age 59 ± 13 years); 45 with LVH only (LVH group; 22 male, 23 female, mean age 62 ± 12 years); and 48 normal controls (control group; 25 male, 23 female, mean age 58 ± 10 years).
Using a standard commercial ultrasound machine (Vivid 5, GE Vingmed, Horten, Norway) with a 2.5-MHz phased array probe, we acquired three apical views (apical four-chamber, two-chamber, and long-axis views) in gray scale and color tissue Doppler modes. Mitral inflow velocities were recorded by using conventional pulsed-wave Doppler echocardiography, in the usual fashion. All images were saved digitally in raw-data format to magneto optical disk for offline analysis.
Echocardiographic data analysis
Left ventricular diameters and wall thicknesses were measured from two-dimensional targeted M-mode echocardiography, using the criteria of the American Society of Echocardiography (14). Fractional shortening was calculated using the standard formula (15), and LV mass was determined by Devereux’s formula (16). Left ventricular hypertrophy was defined as LV mass index (g/m2) >131 g/m2in men and >100 g/m2in women (17). Resting LV end-diastolic, end-systolic volumes and ejection fraction were computed using a modified Simpson’s biplane method. Each representative value was obtained from the average of three measurements.
Strain and strain rate data analysis
Tissue velocity curves were obtained from color tissue Doppler images using a computer (Apple Macintosh G4, Cupertino, California) and standard commercial software (Echopac, GE Vingmed). Peak myocardial early diastolic velocity (Em) and peak myocardial late diastolic velocity (Am) and their ratio were obtained by placing a tissue Doppler sample volume at the septal annulus in the apical four-chamber view.
Myocardial strain is a fundamental quality of tissue that reflects its ability to shorten. Strain and strain rate curves were extracted from an average of three cycles of tissue Doppler imaging data, using an IBM computer and developmental software (Formtest V6.1, GE Vingmed). Strain and strain rate were derived from strain and strain rate curves obtained by placing sample bar (12 mm) on six walls in the three apical views (18,19). Sampling in the midmyocardial layer was performed in each segment and maintained at the same position during the cardiac cycle by manually tracking wall motion, but data were excluded if we were unable to obtain a smooth strain curve or the angle between the scan-line and wall was >20°. Peak strain was defined as the greatest value on the strain curve.
Peak strain (26 ± 4% in controls) is a measure of tissue distortion and a marker of regional systolic function, which has been demonstrated to be able to quantify LV function in acute myocardial infarction (20). Peak strain rate (1.6 ± 0.3 s−1in controls) is an index of the speed of contraction and can be estimated as the spatial derivative of velocities. It is an elegant noninvasive indicator of LV contractility and has been validated in dogs by comparison with peak elastance, a gold standard of LV contractility (21).
Calibrated IB was obtained by comparison of the septal or posterior wall IB intensity with pericardial IB intensity in the parasternal long-axis view. The IB curve for the septum, posterior wall or pericardium was extracted using a computer (Apple Macintosh G4) and standard commercial software (Echopac, GE Vingmed). Measurements were obtained by placing a 9 × 9 pixel sample volume in the basal septum, posterior wall or pericardium in end-diastole. The position of the sample volume was checked and adjusted in each frame to keep the sample volume within the same region during the whole cardiac cycle. Calibrated IB was obtained by subtracting average pericardial IB intensity from average myocardial IB intensity of the septum or posterior wall.
Interobserver and intraobserver variability
Variability in the measurement of peak strain, strain rate, and calibrated IB from a single acquisition was evaluated in 30 subjects (8 from DM group, 8 from DH group, 7 from LVH group and 7 controls) randomly selected from the 186 subjects by two independent observers for interobserver and intraobserver variability. To determine reproducibility, the same observer who was blinded to the former results measured peak strain, strain rate and calibrated IB for each of the selected patients again at a separate time (at least two weeks later). To test interobserver variability, another observer (unaware of patient identity and first observer’s results) analyzed the same patients’ data in the same way.
Values were expressed as a mean ± standard deviation. One-way analysis of variance (ANOVA) with post-hoc test by Bonferroni was used to examine differences among groups. Student’s independent-samples ttest was used to compare the difference between two groups. Data were analyzed using standard statistical software (SPSS Inc., Chicago, Illinois). A p value of <0.05 was considered statistically significant.
Clinical and echocardiographic characteristics
Table 1summarizes the clinical characteristics of the four groups. One-way ANOVA with post-hoc test by Bonferroni showed they had comparable mean ages, heart rates, body weights, smoking histories, and LV systolic and diastolic functions. Patients from the DM group showed a greater prevalence of obesity and hypercholesterolemia than controls, and those with LVH had greater systolic and diastolic blood pressure and LV dimensions and mass than controls. In all patient groups (DM, DH, LVH), Em (5.0 ± 2.4 cm/s; 4.7 ± 1.9 cm/s; 5.6 ± 1.5 cm/s, respectively) and Am (6.4 ± 2.2 cm/s; 6.5 ± 1.7 cm/s; 7.1 ± 1.2 cm/s, respectively) were less than in controls (7.0 ± 2.1 cm/s for Em, 7.7 ± 1.6 cm/s for Am).
There were no significant differences in diabetic duration, types and complications, hemoglobin A1c(HbA1c), blood glucose, lipid profile (except low-density lipoprotein cholesterol), and treatments between DM and DH groups, but creatinine, urea, and low-density lipoprotein cholesterol were significantly greater in the DH group than in the DM group (Table 2). Indices of diabetic control (HbA1cor blood glucose) were not correlated with myocardial function (peak strain, strain rate), or reflectivity (calibrated IB). There were no significant differences in peak strain, strain rate, and calibrated IB in subgroups of patients with or without complications (e.g., nephropathy, retinopathy, and neuropathy). These parameters did not differ according to treatment with and without angiotensin-converting enzyme inhibitors or between patients treated with insulin or oral hypoglycemic drugs.
Comparison of myocardial changes in the DM group, LVH group, or both groups
The results analyzed by one-way ANOVA with post-hoc test by Bonferroni in peak strain, strain rate, and calibrated IB for the four groups are shown in Figure 1. All three patient groups (DM, DH, and LVH) showed significant decreases in peak strain compared with controls (24 ± 3%, 22 ± 3%, 24 ± 3%, all p values ≤ 0.001). Peak strain was significantly less in the DH group than in the DM group (p = 0.03) and the LVH group (p = 0.02). Strain rate was greater in controls (1.6 ± 0.3 s−1) than in DM (1.4 ± 0.3 s−1, p = 0.006), DH (1.3 ± 0.2 s−1, p < 0.001), and LVH (1.4 ± 0.2 s−1; p = 0.005) groups. Moreover, peak strain rate in the DH group was significantly less than in the DM alone (p = 0.01) or LVH (p = 0.01) groups.
Calibrated IB in the septum (−22.6 ± 6.6 dB in controls) and posterior wall (−29.0 ± 6.2 dB) is a marker of myocardial reflectivity. With the exception of the results in the posterior wall of the DM group, the three patient groups (DM, DH, LVH) showed significant increases in calibrated IB both in the septum (−18.6 ± 7.9 dB, −17.1 ± 7.0 dB, −17.2 ± 6.4 dB; p = 0.045, 0.001, 0.001, respectively) and in the posterior wall (−27.0 ± 7.0 dB, −23.8 ± 7.6 dB, −25.0 ± 7.7 dB; p = 1.0, 0.004, 0.046, respectively). There were no significant differences in the septal or posterior wall calibrated IB among the three patient groups.
Interobserver and intraobserver variability
There were no significant differences in peak strain (24.1 ± 2.8%), strain rate (1.5 ± 0.2 s−1), and calibrated IB of the septum (−19.4 ± 7.6 dB) and posterior wall (−27.4 ± 7.3 dB) when these were measured by another observer (23.6 ± 3.3%, 1.4 ± 0.2 s−1, −19.7 ± 6.4 dB, −26.2 ± 5.8 dB, respectively) or remeasured by the same observer (24.0 ± 3.3%, 1.4 ± 0.2 s−1, −19.1 ± 7.4 dB, −26.2 ± 5.8 dB, respectively). Mean absolute differences in peak strain, strain rate, calibrated IB of the septum and posterior wall were 1.6 ± 1.2% (range 0.1% to 4.6%), 0.1 ± 0.1 s−1(range 0 to 0.3 s−1), −3.1 ± 2.8 (range −0.3 dB to −9.5 dB), and −3.2 ± 2.7 (range −0.1 dB to −9.7 dB) between the two measurements by the same observer and were 1.8 ± 1.2% (range 0 to 4.3%), 0.1 ± 0.1 s−1(range 0.0 to 0.3 s−1), −3.3 ± 2.7 dB (range −0.2 dB to −9.3 dB), −3.7 ± 2.7 dB (range −0.1 dB to −10.0 dB) between the two observers.
Diabetic myocardial changes may be due to metabolic derangements (22), microvascular disease (23), and myocardial fibrosis (24). The most prominent histopathologic findings in diabetic patients without CAD and hypertension are myocellular hypertrophy and myocardial fibrosis (25), implying changes in both myocardial function and structure. As hypertension and LVH frequently coexist with diabetes, we sought in this study to compare myocardial functional and structural changes in diabetic patients with or without LVH with those in nondiabetic subjects with or without LVH, in order to identify the role of diabetes alone in relation to myocardial dysfunction, after exclusion of patients with significant coronary disease as defined by stress echocardiography.
Reduction in myocardial contractility
A number of plausible explanations have been proposed to account for reduced myocardial contractility in diabetes. Chronic abnormalities in myocardial carbohydrate and lipid metabolism due to insulin deficiency may result in reduced adenosine triphosphatase activity, decreased ability of the sarcoplasmic reticulum to take up calcium (26)and an intracellular accumulation of toxic fatty acid intermediates (22). These in turn may lead to adenosine triphosphate depletion, changes in calcium homeostasis and increased myocardial oxygen consumption, and may produce a focal, progressive loss of myofibrils, transverse tubules and sarcoplasmic reticulum, and separation of the fasciae adherens at the intercalated disk within myocytes (27), causing myocyte hypertrophy, loss and replacement of fibrosis, and resulting in deleterious effects on myocardial contractility.
Despite epidemiologic observations of the greater frequency of heart failure in diabetic subjects, and credible explanations for LV dysfunction, some studies in diabetic patients without overt evidence of heart disease have demonstrated normal contraction at rest. In these circumstances, the contractile response during exercise was abnormal (28), suggesting loss of contractile reserve in the early phase of diabetic heart disease. In this situation, resting changes may be too subtle to be identified with load-dependent indicators, such as ejection fraction, and require the application of sensitive techniques. This study demonstrated that peak strain and strain rate were significantly reduced in patients with diabetes mellitus and that these changes were analogous to those associated with LVH—which is recognized as an important cause of myocardial dysfunction. Moreover, the changes due to diabetes and LVH appear to be synergistic, as evidenced by the significantly lower myocardial contractility measured by peak strain and strain rate in the DH group; these findings are consistent with those of previous studies that show the combination of diabetes and hypertension to have an adverse effect on the myocardium (24). Left ventricular hypertrophy in the hearts of some patients with diabetes may be mainly secondary to diabetes rather than hypertension, supported by seven patients in the DH group who had no history of hypertension.
Increase in myocardial reflectivity
Previous work has shown positive associations between heart weight and total fibrosis in patients with diabetes alone and with both diabetes and LVH (24). Collagen is the primary determinant of echocardiographic scattering in myocardial tissue and there is quite good correlation between collagen deposition and backscatter magnitude (29,30). The present study showed calibrated IB was significantly increased in the three patient groups, which confirms the previous findings (24,31)and suggests that diabetes is as likely as LVH to be associated with myocardial fibrosis, and this may be an important factor in early diabetic heart disease. These results are similar to those from a study that revealed greater replacement of myocardium by fibrosis in rats with both diabetes and hypertension than in rats with diabetes alone (32).
Several limitations are inherent in this observational study. Most patients were on medication, and the average HbA1cwas 7% to 8%, suggesting imperfect but moderately successful control of blood sugar levels. A greater spectrum of control might have facilitated elucidation of the relationship between myocardial dysfunction and glycemic control. Similarly, a greater spectrum of patients might afford better insight into the relationship of dysfunction with duration of diabetes and also with medical therapy. Finally, exclusion of an ischemic contribution was mainly based on a normal result of dobutamine echocardiography, although 13 patients had both normal dobutamine echocardiography and coronary angiography. Clearly this does not exclude the possibility of CAD because of the possibility of false-negative results, but these are generally associated with mild disease, and it seems unlikely that a major ischemic contribution was present. This limitation is unavoidable because it would be difficult to justify coronary angiography in asymptomatic diabetic patients with normal dobutamine echocardiograms on ethical grounds.
The results of this study indicate that myocardial structure (increased calibrated IB) and function (decreased peak strain and strain rate) are altered before the development of myocardial systolic dysfunction in the hearts of patients with diabetes, and these alterations potentially may be related to the subsequent development of overt diabetic cardiomyopathy. Although similar to changes caused by LVH, these changes are independent of LVH and support a synergistic relationship between diabetes and LVH as a cause of diabetic heart disease.
☆ This study was supported in part by grant 993601 from the National Health and Medical Research Council of Australia.
- peak myocardial late diastolic velocity
- analysis of variance
- coronary artery disease
- DH group
- diabetic patients with left ventricular hypertrophy
- DM group
- diabetic patients without left ventricular hypertrophy
- peak myocardial early diastolic velocity
- hemoglobin A1c
- integrated backscatter
- left ventricular
- left ventricular hypertrophy
- LVH group
- nondiabetic patients with left ventricular hypertrophy
- Received April 26, 2002.
- Revision received August 26, 2002.
- Accepted November 1, 2002.
- American College of Cardiology Foundation
- Grundy S.M.,
- Benjamin I.J.,
- Burke G.L.,
- et al.
- Mehta S.,
- Nuamah I.,
- Kalhan S.
- Perez J.E.,
- McGill J.B.,
- Santiago J.V.,
- et al.
- Bello V.D.,
- Talarico L.,
- Picano E.,
- et al.
- Jamal F.,
- Strotmann J.,
- Weidemann F.,
- et al.
- Gotte M.J.W.,
- van Rossum A.C.,
- Twisk J.W.R.,
- et al.
- UK Prospective Diabetes Study Group
- Sahn D.J.,
- DeMaria A.,
- Kisslo J.,
- Weyman A.
- Schiller N.B.,
- Shah P.M.,
- Crawford M.,
- et al.
- Devereux R.B.,
- Reiche N.
- Edvardsen T.,
- Skulstad H.,
- Aakhus S.,
- et al.
- Edvardsen T.,
- Gerber B.L.,
- Garot J.,
- et al.
- Greenberg N.L.,
- Firstenberg M.S.,
- Castro P.L.,
- et al.
- Nitenberg A.,
- Valensi P.,
- Sachs R.,
- et al.
- van Hoeven K.H.,
- Factor S.M.
- Picano E.,
- Pelosi G.,
- Marzilli M.,
- et al.
- Mimbs J.W.,
- O’Donnell M.,
- Bauwens D.,
- et al.
- Lucarini A.R.,
- Talarico L.,
- Di Bello V.,
- et al.