Quantitative assessment of myocardial function is of great importance in the diagnosis of heart diseases. Standard echocardiography is commonly performed on both humans and small animals to non-invasively assess myocardial function. Several 2D and M-mode measurements (such as systolic left ventricular diameter and index volume, and fractional shortening, %FS) are often used as indices of systolic myocardial performance. Similarly, several conventional Doppler indices (such as E:A ratio of mitral inflow) are commonly used to detect impaired diastolic myocardial function. Tissue Doppler imaging (TDI) is a more recent ultrasound technique allowing quantification of global and regional myocardial function by measurement of myocardial velocities in real time.1 TDI offers a new noninvasive and quantitative analysis of the two intrinsic motions of the myocardium during the whole cardiac cycle, both the radial and the longitudinal.2-4 Karl Isaaz, a French cardiologist, was the first to record the left ventricular free wall (LVFW) velocities in human patients using standard ultrasound equipment.5 Since then, the technique has much improved with the development of specific software and the use of color Doppler imaging.6 TDI has been increasingly investigated in human cardiology and also in veterinary cardiology since the past 10 years, providing sensitive information on myocardial motion, and therefore improving diagnostic abilities of myocardial dysfunction.
TDI Principles and TDI Modes
In order to display the low velocity-high amplitude Doppler signals of the myocardium and to suppress the high velocity-low amplitude Doppler signals of blood flow, specific adjustments are required for TDI examination such as suppressing the high-pass filters and decreasing the gain setting.1
Three TDI modes are available:1 the pulsed-wave mode, the two-dimensional (2D) color mode and the color M-mode. The pulsed-wave TDI mode provides information on myocardial movements through a single sample gate, which is placed within the myocardial wall thickness. With the color M-mode, myocardial velocities are analyzed along a selected single scan line, which is placed in the same manner as for conventional transventricular M-mode. M-mode TDI tracings show on the same image both systolic and diastolic velocities within the entire wall thickness. As with conventional color Doppler system, myocardial velocities towards the transducer are encoded in red, and those away from the transducer in blue. An off-line analysis using a specific software allows quantification of myocardial velocities in the different aligned layers with a high temporal resolution and also a high signal-to-noise ratio. Using 2D color TDI mode, real time color Doppler is superimposed on the gray-scale and the Doppler receive gain is adjusted to maintain optimal coloring of the myocardium. After storage of 2D color TDI loops, an off-line analysis of myocardial velocities is performed in one or several segments using a specific software. One of the main advantages of 2D color TDI mode over the two others is its ability to simultaneously quantify myocardial velocities in several segments, thereby allowing assessment of intra- and interventricular myocardial synchronization.7
Normal TDI Examination and Variability
Normal Velocity Profiles in the Dog and in the Cat
Radial8-13 and longitudinal8,10-13 LVFW velocities may be quantified in small animals using the right parasternal short axis view and the left apical 4-chamber view, respectively. After a short isovolumic contraction phase, all radial and longitudinal velocity profiles include one positive systolic wave (S), and after a short isovolumic relaxation phase, two diastolic negative waves (E and A, respectively in early and late diastole). Biphasic shifts may be recorded during early diastole as well as during the isovolumic relaxation and isovolumic contraction phases.9 Fusion of the two negative diastolic waves E and A into one negative diastolic wave EA is often observed in the cat owing to rapid heart rate, and this represents the main limitation of the TDI technique in this species.13
In the cat9,11,13 as well as in the dog,10,12 normal radial myocardial motion is characterized by nonuniformity, with myocardial layers moving more rapidly in the sub-endocardium than in the sub-epicardium during the whole cardiac cycle, thus defining intramyocardial radial velocity gradients (MVG) in both systole and diastole. Normal longitudinal myocardial motion is also characterized by nonuniformity, with myocardial velocities decreasing from the base to the apex, thus defining longitudinal MVGs in both systole and diastole. A physiologic nonhomogeneity in the longitudinal myocardial motion has also been demonstrated in the normal cat between the interventricular septum and the LVFW, with higher early diastolic velocities, acceleration, and deceleration in the former compared with the latter.14 Lastly, right ventricular myocardial (RVM) velocities have been shown to be higher in the basal than in the apical segments, thus defining a significant right base-to-apex MVG, with RVM velocities higher than the LVFW velocities of the corresponding segment at each phase of the cardiac cycle.15
Intra-Observer Variability (Repeatability and Reproducibility)
The intra-observer within-day (repeatability) and between-day (reproducibility) variability of LVFW myocardial velocities has been shown to be correct to good for most TDI variables in the awake dog12 and cat.13,16 In both species, as observed in human patients,17 the highest variability is observed at the apex. Training has been shown to improve the variability of the TDI technique: for example, in a first protocol, the between-day CV values of longitudinal basal LVFW velocities obtained for a 2-year trained observer were relatively high, i.e., 27%, 19.0% and 17.8% and 50.9% for S, E and A waves, respectively.12 After 2 additional years of training, the between-day CV values were much lower, i.e., all < 15% (10.3%, 5.6% and 10.9%, respectively).17
Other Factors of Variation
A breed effect has been documented for several systolic and diastolic TDI variables in the cat and in the dog.10,11 In the dog,10 a positive correlation has been shown between heart rate and longitudinal S wave at the base. A similar relationship has been reported in the cat for several TDI variables.11 TDI studies have also revealed a progressive decrease in systolic and early diastolic velocities and an increase in late diastolic velocities as age-related changes in healthy human subjects.16 Lastly, one study performed on healthy dogs showed that anesthesia significantly decreased both radial and longitudinal myocardial velocities, up to 60% compared to values assessed in awake animals.12
TDI Applications: Early and Sensitive Diagnosis of Myocardial Dysfunction
TDI offers a non-invasive and sensitive quantitative analysis of regional radial and longitudinal myocardial motion. This technique may therefore be used to accurately investigate myocardial dysfunction associated with heart diseases, thus providing new insights in the understanding of their pathophysiology. For example, diastolic dysfunction has traditionally been thought to be the only abnormality in cats with hypertrophic cardiomyopathy (HCM). One study showed that systolic dysfunction is an additional component of myocardial alteration characterized by a decrease in longitudinal systolic velocities and gradients (despite normal or increased fractional shortening) and the presence of high post-systolic shortening waves.18 Another recent study confirmed these results, demonstrating a systolic impairment along the longitudinal axis of the LVFW in cats with HCM.14 Similarly, dilated cardiomyopathy (DCM) is an idiopathic myocardial disorder characterized by primary impairment of systolic function. One study performed by our group demonstrated that spontaneous canine DCM is commonly associated with diastolic LVFW alteration, characterized by significantly decreased radial and longitudinal E waves, associated with an inverse TDI E/A ratio in several dogs.19
The above diastolic and systolic myocardial alterations may sometimes be present before overt conventional echo-Doppler abnormalities. Another important application of TDI (in research or clinical settings) is therefore the early detection of myocardial lesions that are equivocal or even not apparent using conventional ultrasound techniques. Several studies have shown that TDI is a more sensitive imaging technique than conventional Doppler echocardiography for detecting myocardial lesions associated with ischemia,20,21 heart transplant disorders,22 and cardiomyopathies in both animal models and humans23-29. Using a dog model of DCM, our group has demonstrated that TDI is more sensitive than conventional echocardiography in detecting preclinical regional myocardial abnormalities before the occurrence of left ventricular dilation and overt systolic dysfunction.26,27 In another study involving cats with myocardial hypertrophy (related to spontaneous HCM or systemic arterial hypertension), those with non-hypertrophied LVFW as assessed by M-mode examination (and presenting only a segmental myocardial hypertrophy localized in the subaortic interventricular septum) displayed systolic and diastolic LVFW dysfunction, thus proving that functional myocardial alteration was more widespread than predicted by conventional M-mode echocardiography and that TDI is capable of detecting segmental functional changes in apparently non-altered wall segments.18 Similarly, in a feline model of HCM, TDI has been shown to consistently detect LVFW dysfunction despite the absence of myocardial hypertrophy in affected males or carrier females.24 A study on canine DCM also revealed that one-third of the dogs with right systolic TDI alterations did not show any right heart dilatation on conventional echocardiography and none of those with TDI LVFW diastolic dysfunction showed an abnormal mitral inflow pattern on conventional Doppler.20 Lastly, the diagnostic value of several echo-Doppler and TDI variables have been compared in dogs with pulmonary arterial hypertension (PAH).30 In this study, TDI provided the most effective (highest sensitivity and specificity) predictors of PAH. This study also demonstrated that both systolic and diastolic right-sided myocardial dysfunction can occur with mild PAH.
Another important application of TDI is the accurate assessment of treatment effect on regional myocardial function. For example, our group has recently used the TDI technique to demonstrate the beneficial regional myocardial effect of non-cultured skeletal muscle cell transplantation in an animal model of non-ischemic DCM.31,32
1. Chetboul V. J Vet Cardiol 2002;4:7.
2. Uematsu M, et al. J Am Coll Cardiol 1995;26:217.
3. Jones CJH, et al. Brit heart J 1990;63:215.
4. Derumeaux G, et al. Arch Mal Coeur Vaiss 1997;90:773.
5. Isaaz K, et al. Am J Cardiol 1989;64:66.
6. Sutherland GR, et al. J Am Soc Echocardiogr 1994;7:441.
7. Estrada A, et al. J Vet Cardiol, 2006;8:129.
8. Gavaghan BJ, et al. Am J Vet Res 1999;60:1478.
9. Koffas H, et al. J Vet Intern Med 2003;17:510.
10. Chetboul V, et al. Am J Vet Res 2005;66:953.
11. Chetboul V, et al. Am J Vet Res 2006;67:250.
12. Chetboul V, et al. Am J Vet Res 2004;65:909.
13. Chetboul V, et al. Am J Vet Res 2004;65:566.
14. Koffas H, et al. J Vet Intern Med 2006;20:65.
15. Chetboul V, et al. J Vet Intern Med 2005;19:837.
16. Simpson KE, et al. Vet Radiol Ultrasound 2007;48:58.
17. Nikitin NP, et al. J Am Soc Echocardiogr 2003;16:906.
18. Carlos Sampedrano C, et al. J Vet Intern Med 2006;20:1106.
19. Chetboul V, et al. J Vet Intern Med 2007;21:719.
20. Derumeaux G, et al. Circulation 2000;101:1390.
21. Derumeaux G, et al. Circulation 2001;103:589.
22. Derumeaux G, et al. Arch Mal Coeur Vaiss 1998;91:1255.
23. Nagueh SF, et al. Circulation 2000;102:1346.
24. Chetboul V, et al. J Vet Intern Med 2006;20:640.
25. Chetboul V, et al. Vet Radiol Ultrasound 2006;47:295.
26. Chetboul V, et al. Am J Vet Res 2004;65:1335.
27. Chetboul V, et al. Eur Heart J 2004;25:1934.
28. Chetboul V, et al. J Am Vet Med Assoc 2004;225:1877.
29. Nagueh SF, et al. Circulation 2001;104:128.
30. Serres V, et al. J Vet Intern Med 2007;21:1280.
31. Borenstein N, et al. J Card Surg 2006;21:50.
32. Borenstein N, et al. Eur J Cardiothorac Surg 2007;31:444.