- Email: [email protected]
Left Ventricular Tei Index in Children: Comparison of Tissue Doppler Imaging, Pulsed Wave Doppler, and M-Mode Echocardiography Normal Values
Left Ventricular Tei Index in Children: Comparison of Tissue Doppler Imaging, Pulsed Wave Doppler, and M-Mode Echocardiography Normal Values Wei Cui, MD, and David A. Roberson, MD, Oak Lawn, Illinois
The Tei index has been found to be useful for analyzing systolic and diastolic global ventricular function in a wide variety of congenital and acquired cardiac abnormalities. However, there are some discrepancies between reports as to the normal values for the Tei index obtained by the different echocardiographic techniques and by different investigators. We conducted a prospective study to determine the normal range of left ventricular Tei index (LVTX) values in a broad sample of children using tissue Doppler imaging, pulsed wave Doppler, and Mmode echocardiography. In all, 289 children with normal echocardiogram findings (age 1 day-18 years, body surface area 0.08-2.4 m2, heart rate 46-182/min) were studied. The LVTX was calculated by
The Tei index has become a widely used echocar-
diographic parameter for the assessment of global systolic and diastolic function in adults and children with a wide variety of congenital and acquired cardiac abnormalities.1-7 However, there are discrepancies between reports regarding the normal values in children, and variability in normal values depending on which echocardiographic modality is used to measure the Tei index.3,8-10 We undertook this prospective study to determine the normal values for the left ventricular (LV) Tei index (LVTX) measured from tissue Doppler imaging (TDI), pulsed wave Doppler (PWD), and M-mode echocardiography (MME). A cohort of children with normal echocardiogram results who spanned the range of ages, body sizes, and heart rates encountered in the pediatric population were studied. The normal values for each of the 3 methods were compared to determine any difference in the normal values depending on which technique was used. The effects of age, body surface area (BSA), and heart rate on the LVTX From the Heart Institute for Children, Hope Children’s Hospital. Reprint requests: David A. Roberson, MD, The Heart Institute for Children, Hope Children’s Hospital, 4440 W 95th St, Oak Lawn IL 60453 (E-mail: [email protected]). 0894-7317/$32.00 Copyright 2006 by the American Society of Echocardiography. doi:10.1016/j.echo.2006.06.006
1438
all 3 methods in each patient during a single echocardiographic examination. The normal LVTX values (mean ⴞ SD) for the 3 techniques were: LVTX-Doppler tissue imaging ⴝ 0.38 ⴞ 0.06; LVTX-pulsed wave Doppler ⴝ 0.36 ⴞ 0.07; and LVTX-M-mode echocardiography ⴝ 0.29 ⴞ 0.08. LVTX-Doppler tissue imaging and LVTX-pulsed wave Doppler values were only slightly but statistically significantly different (P < .05). LVTXM-mode echocardiography values were consistently and significantly less than those obtained by both of the other two methods (P < .01, respectively). The effects of age, body surface area, and heart rate were not clinically significant. These results are similar but not identical to those from prior studies. (J Am Soc Echocardiogr 2006;19:1438-1445.)
were analyzed. Lastly, our results were compared with those from previous studies.3,8-10 METHODS Study Design Tei index values obtained from TDI, PWD, and MME were each measured from the same normal echocardiogram result to determine the normal values for each technique. LVTX measurements were made at times of stable heart rate with the R-R interval varying by 40 milliseconds or less between measurements. Patients were grouped according to age as in prior studies.8 BSA was calculated and recorded in each case. The heart rate was calculated from the R-R interval at the time of LVTX measurements. Normal values for each technique were compared to determine whether the values obtained differed for the 3 techniques. Results for different age groups, BSAs, and heart rates were compared to determine any effect of these parameters on the LVTX obtained from all 3 methods. Patient Features The study group consisted of 289 pediatric patients with normal echocardiogram findings. Patient features are summarized in Table 1. The patient ages ranged from 1 day to 18 years. BSA spanned from 0.08 to 2.4 m2. The minimum heart rate was 46/min and the maximum heart rate was 182/min.
Journal of the American Society of Echocardiography Volume 19 Number 12
Cui and Roberson 1439
Table 1 Demographics in study patients n Male Age Range BSA, m2 Range HR, beats/min Range
<1 mo
1 mo-1 y
1-6 y
6-10 y
10-14 y
14-18 y
Total
46 23 7.3 ⫾ 7.3 d 1-29 d 0.187 ⫾ 0.06 0.08-0.40 139 ⫾ 14 113-182
35 16 3.0 ⫾ 2.3 mo 1-11 mo 0.28 ⫾ 0.07* 0.12-0.41 147 ⫾ 19* 90-181
50 28 2.8 ⫾ 1.2 y 1-5 y 0.61 ⫾ 0.11† 0.41-0.90 109 ⫾ 21† 71-162
45 20 7.3 ⫾ 1.2 y 6-9 y 0.96 ⫾ 0.18† 0.59-1.30 89 ⫾ 12† 55-115
44 22 11.7 ⫾ 1.0 y 10-13 y 1.45 ⫾ 0.29† 0.88-2.40 75 ⫾ 12† 46-107
69 40 15.5 ⫾ 1.1 y 14-18 y 1.74 ⫾ 0.21† 1.33-2.40 72 ⫾ 13 46-109
289 149 7.1 ⫾ 6.2 y 1 d-18 y 0.95 ⫾ 0.62 0.08-2.40 101 ⫾ 33 46-182
BSA, Body surface area; HR, heart rate. *P ⬍ .05; †P ⬍ .01 compared with preceeding age group. Data expressed as mean ⫾ SD.
Equipment Echocardiograms were performed by full-time clinical pediatric echocardiographic technicians and cardiology fellows or attending physicians. Three echocardiographic systems were used (Sonos 5500 and 7500, Philips Medical Systems, Andover, Mass; and Sequoia 512, Siemens Medical Solutions, Malvern, Pa). All images were stored digitally for further analysis and review (Xcelera system, Philips Medical Systems).
edge to the trailing edge of the left ventricular outflow tract (LVOT) PWD tracing (Figure 1, B). Both TDI and PWD recordings were made from the standard apical transducer position. LVTX-MME a component was measured from mitral valve closure to the subsequent mitral valve opening on the mitral valve MME tracing. The LVTX-MME b component was measured from aortic valve opening to aortic valve closure on the aortic valve MME tracing. MME recordings were obtained from the standard parasternal long axis view (Figure 1, C ).
Echocardiographic Studies Complete 2-dimensional, Doppler, color Doppler, and MMEs were performed on all patients. We include apical 4-chamber TDI of the mitral annulus in our standard clinical echocardiogram protocol. PWD of the mitral inflow and LV outflow, and MME of the mitral and aortic valves are also components of a standard complete echocardiogram. Echocardiograms were ordered for clinical purposes by the clinical attending physicians or house staff. For the purpose of this study, only echocardiograms with an official reading of completely normal study or completely normal study except for patent foramen ovale with a diameter of less than or equal to 2 mm and trivial left-to-right shunt were accepted for analysis. Tei Index Calculation The Tei index was calculated as previously described.3,8-14 Specific details of our measurement techniques were as follows. For all 3 methods the a component equals the sum of isovolumic contraction time plus ejection time plus isovolumic relaxation time. The b component is equal to the left ventricular ejection time. The LVTX is calculated as the difference of a minus b divided by b. The LVTX-TDI a component was measured from the trailing edge of the mitral annular A’ wave to the leading edge of the subsequent TDI mitral annular early diastolic (E’) wave. The LVTX-TDI b component was measured from the leading edge to the trailing edge of the TDI mitral annular systolic (S) wave (Figure 1, A). LVTX-PWD a component measurements were made from the trailing edge of the PWD late mitral A wave to the leading edge of the subsequent PWD early mitral E wave. The b component for the LVTX-PWD was measured from the leading
Statistical Analysis Continuous data are reported as the mean ⫾ SD. Discrete data are reported as frequencies and percentages. Oneway analysis of variance and analysis of variance for repeated measures analysis with post hoc Tukey test were used to determine significant differences of continuous data between groups. The effects of age, heart rate, and BSA on LVTX were evaluated by multiple correlation and linear regression analysis. Statistical significance was declared when the computed P value was less than .05. All statistical tests were 2-sided and all analyses were performed using software (SAS for Windows, SAS Inc, Cary, NC). Reproducibility Intraobserver variability was assessed in 22 participants by repeating the measurements on two occasions 5 days apart. To test the interobserver variability, the measurements were performed from digitally stored cycles by a second observer who was unaware of the prior results. The variability was calculated as the mean percentage error, derived as the difference between the two sets of measurements, divided by the overall mean of the observed values by the two measurements. Human Research This study complies with our institutional policies for human research and patient confidentiality.
1440 Cui and Roberson
Journal of the American Society of Echocardiography December 2006
Figure 1 A, Left ventricular (LV) Tei index (LVTX) calculation using tissue Doppler imaging (TDI). The a component (isovolumic contraction time ⫹ ejection time ⫹ isovolumic relaxation time) is measured from trailing edge of late diastolic TDI mitral annular A wave to leading edge of subsequent early diastolic TDI mitral annular E wave. The b component is measured from leading edge of systolic TDI mitral annular S wave to trailing edge of systolic TDI mitral annular S wave. LVTX formula is then calculated by difference of: a ⫺ b/b. B, LVTX calculation using tissue pulsed wave Doppler (PWD). The a component is measured from trailing edge of late diastolic transmitral PWD flow A wave to leading edge of subsequent early diastolic transmitral PWD flow E wave. The b component is measured from leading to trailing edge of left ventricular outflow systolic PWD tracing. LVTX formula is then calculated by difference of: a ⫺ b/b.
Journal of the American Society of Echocardiography Volume 19 Number 12
Cui and Roberson 1441
Figure 1 C, Left ventricular Tei index (LVTX) calculation using M-mode echocardiography (MME). The a component is from mitral valve closure to following mitral valve opening. The b component is from aortic valve opening to closing.
Figure 2 Left ventricular Tei index (LVTX) normal values for tissue Doppler imaging (TDI), pulsed wave Doppler (PWD), and M-mode echocardiography (MME) methods expressed as mean ⫾ SD.
RESULTS Normal LVTX values for the 3 techniques expressed as mean ⫾ SD (Figure 2) were: LVTX-TDI ⫽ 0.38 ⫾ 0.06; LVTX-PWD ⫽ 0.36 ⫾ 0.07; and LVTX-MME ⫽ 0.29 ⫾ 0.08. There was a small but statistically significant difference in LVTX-TDI versus LVTXPWD (P ⬍ .05). LVTX-MME was consistently and significantly less than LVTX-TDI and LVTX-PWD
(P ⬍ .01, respectively). Z scores for the 3 techniques are presented in Figure 3. Results for the normal values of LVTX-TDI, LVTXPWD, and LVTX-MME for various age groups are summarized in Table 2. There were some very small but statistically significant differences between some of the age groups for LVTX-TDI results. These included slightly lower LVTX-TDI normal values in age group 1 month to 1 year versus both age groups 10 to 14 years and 14 to 18 years (P ⬍ .05, respectively). There was also a slight difference in the age group 1 to 6 years LVTX-TDI being slightly less than the age group 14 to 18 years (P ⬍ .05). There was also a slight difference for LVTX-PWD results between patients age 1 month to 1 year versus both age groups 10 to 14 years and 14 to 18 years (P ⬍ .05, respectively). However, there was no significant association between LVTX by any of the 3 methods and age or heart rate after controlling for the effect of BSA. There was a statistical association between BSA and both LVTXTDI and LVTX-PWD (P ⬍ .01). R2 values of only 0.0579 and 0.0573, respectively, indicate only 5.8% of the variation in LVTX-TDI and 5.7% of the variation in LVTX-PWD can be explained by variation in BSA. Intraobserver and interobserver variability was low with measurement intraobserver variability of 3.0% ⫾ 2.0% and interobserver variability of 3.4% ⫾ 2.0%.
Journal of the American Society of Echocardiography December 2006
1442 Cui and Roberson
Figure 3 Left ventricular Tei index (LVTX) normal values for tissue Doppler imaging (TDI), pulsed wave Doppler (PWD), and M-mode echocardiography (MME) methods expressed as z scores. Table 2 Tei index for left ventricle derived from different methods TDI a
⬍1 mo
0.25 ⫾ 0.02 (0.24-0.26) 1 mo-1 y 0.25 ⫾ 0.03 (0.24-0.26) 1-6 y 0.34 ⫾ 0.04 (0.32-0.34) 6-10 y 0.36 ⫾ 0.03 (0.35-0.37) 10-14 y 0.39 ⫾ 0.02 (0.38-0.40) 14-18 y 0.39 ⫾ 0.03 (0.39-0.40) Overall 0.34 ⫾ 0.07 (0.33-0.35)
PWD
MME
b
Tei index
a
b
Tei index
a
b
Tei index
0.18 ⫾ 0.02 (0.18-0.19) 0.19 ⫾ 0.02 (0.18-0.20) 0.24 ⫾ 0.03 (0.23-0.25) 0.27 ⫾ 0.02 (0.26-0.27) 0.28 ⫾ 0.02 (0.27-0.28) 0.28 ⫾ 0.02 (0.28-0.29) 0.25 ⫾ 0.05 (0.24-0.25)
0.37 ⫾ 0.06 (0.35-0.40) 0.35 ⫾ 0.09‡§ (0.32-0.39) 0.36 ⫾ 0.04‡ (0.35-0.38) 0.37 ⫾ 0.04 (0.36-0.39) 0.40 ⫾ 0.07 (0.38-0.42) 0.40 ⫾ 0.05 (0.38-0.41) 0.38 ⫾ 0.06*† (0.37-0.39)
0.25 ⫾ 0.03 (0.24-0.25) 0.26 ⫾ 0.03 (0.24-0.25) 0.31 ⫾ 0.03 (0.30-0.32) 0.35 ⫾ 0.03 (0.34-0.36) 0.37 ⫾ 0.03 (0.36-0.38) 0.38 ⫾ 0.04 (0.37-0.39) 0.32 ⫾ 0.06 (0.31-0.33)
0.18 ⫾ 0.02 (0.18-0.19) 0.19 ⫾ 0.02 (0.18-0.19) 0.23 ⫾ 0.02 (0.22-0.23) 0.26 ⫾ 0.02 (0.25-0.26) 0.27 ⫾ 0.02 (0.26-0.28) 0.27 ⫾ 0.03 (0.27-0.28) 0.24 ⫾ 0.04 (0.23-0.24)
0.36 ⫾ 0.11 (0.32-0.39) 0.32 ⫾ 0.07‡§ (0.30-0.35) 0.36 ⫾ 0.04 (0.34-0.37) 0.35 ⫾ 0.05 (0.33-0.37) 0.38 ⫾ 0.07 (0.35-0.40) 0.39 ⫾ 0.07 (0.37-0.40) 0.36 ⫾ 0.07† (0.35-0.37)
0.24 ⫾ 0.02 (0.24-0.25) 0.25 ⫾ 0.02 (0.24-0.26) 0.30 ⫾ 0.03 (0.29-0.31) 0.33 ⫾ 0.03 (0.32-0.34) 0.36 ⫾ 0.03 (0.35-0.37) 0.37 ⫾ 0.04 (0.36-0.38) 0.32 ⫾ 0.06 (0.31-0.32)
0.19 ⫾ 0.02 (0.18-0.19) 0.20 ⫾ 0.03 (0.19-0.21) 0.24 ⫾ 0.03 (0.23-0.25) 0.26 ⫾ 0.02 (0.25-0.26) 0.28 ⫾ 0.02 (0.27-0.29) 0.29 ⫾ 0.03 (0.28-0.30) 0.25 ⫾ 0.05 (0.24-0.25)
0.30 ⫾ 0.09 (0.27-0.33) 0.27 ⫾ 0.09 (0.23-0.31) 0.28 ⫾ 0.05 (0.26-0.30) 0.28 ⫾ 0.07 (0.26-0.31) 0.29 ⫾ 0.07 (0.27-0.32) 0.29 ⫾ 0.09 (0.26-0.32) 0.29 ⫾ 0.08 (0.28-0.30)
a, isovolumic contraction time plus ejection time plus isovolumic relaxation time; b, ejection time; M, month; MME, M-mode echocardiography; PWD, pulse waved Doppler; TDI, tissue Doppler imaging; Y, years. *P ⫽ .0007 compared with pulsed Doppler-derived Tei index. †P ⬍ .0001 compared with M-mode-derived Tei index. ‡P ⬍ .05 compared with 14-18 y age group. §P ⬍ .05 compared with 10-14 y age group. Data expressed as mean ⫾ SD (95% confidence interval).
DISCUSSION These LVTX normal values can be applied to the entire spectrum of the pediatric population with no clinically significant effects of age, heart rate, and BSA throughout this group of patients. Although there were some slight statistical differences between LVTX results as a result of the effect of BSA, this difference of approximately 0.02 is so small that it seems negligible in the clinical setting. Once we controlled for the slight
effect of BSA, we did not find any effect of age on the LVTX in the pediatric population, in contrast to the findings of Spencer et al,14 who demonstrated some age-dependent changes of LVTX in adult patients. It was not surprising to find some differences between the LVTX values for the 3 methods, because they actually measure different time interval parameters for the a and b components of the LVTX. Typical differences in the timing and duration of a and b measurements are illustrated in the composite
Journal of the American Society of Echocardiography Volume 19 Number 12
Cui and Roberson 1443
Figure 4 Composite echocardiograms constructed from one echocardiogram to illustrate typical differences in timing and duration of a and b components of left ventricular (LV) Tei index (LVTX) for tissue Doppler imaging (TDI) (top), pulsed wave Doppler (PWD) (middle), and M-mode echocardiography (MME) (bottom). Events were timed with respect to electrocardiogram and TDI a and b components. A, Measurement of a component of LVTX by 3 methods. TDI a starts and ends later than PWD a and durations of a are similar for both. MME a starts at same time as TDI a but its duration is shorter than TDI a and PWD a. B, Measurement of b component of LVTX by 3 methods. TDI b starts and ends earlier than PWD b and durations of b are similar for both. MME b starts at same time as TDI b but its duration is longer than TDI b and PWD b. A, Mitral valve PWD A wave; A’, mitral annulus late diastolic myocardial velocity; AoV, aortic valve; E, mitral valve PWD E wave; E’, mitral annulus early diastolic myocardial velocity; LVOT, LV outflow tract; MV, mitral valve; R, R wave of QRS complex.
1444 Cui and Roberson
Figure 5 Between studies comparison of left ventricular Tei index (LVTX) normal values for tissue Doppler imaging (TDI), pulsed wave Doppler (PWD), and Mmode echocardiography (MME). Data are expressed as mean ⫾ SD. n, Number of patients in each study. Five studies compared are current study (Cui) and those of Eidem et al,8 Gaibazzi et al,3 Tham and Silverman,9 and Harada et al,10 respectively.
echocardiograms constructed for Figure 4. Figure 4, A, demonstrates that the TDI a begins after the PWD a begins and ends after the PWD a ends. The MME a begins at the same time as the TDI a but ends before both the TDI a or PWD a ends. Therefore, the a measurements are similar for TDI and PWD, but shorter for MME. Figure 4, B, illustrates that the TDI b wave begins slightly before the PWD b begins and that the TDI b ends slightly before the PWD b ends. The MME b begins at the same time as the TDI b but ends after both the TDI b and PWD b, resulting in a longer MME b versus both TDI b and PWD b. In other words, TDI and PWD a and b parameters have slightly different timing but similar duration, resulting in similar LVTX-TDI and LVTX-PWD. However, for MME, the a is shorter and the b is longer than the TDI and PWD parameters. This causes the LVTXMME to be smaller than the LVTX-TDI or LVTXPWD. There is extensive overlap between our results and those of other investigators (Figure 5). LVTXTDI results were similar to those obtained by Harada et al10 and Gaibazzi et al.3 The widest range of results between studies, albeit with significant overlap of the normal value range, was obtained by the PWD technique.3,8-10 A possible source of the minor discrepancies between studies may be difficulty in precisely measuring the termination point of mitral valve inflow (ie, the end of the mitral valve A wave)
Journal of the American Society of Echocardiography December 2006
by PWD because of low velocity noise around the baseline at this time in the cardiac cycle. In addition, LVTX-PWD measurement sometimes requires the use of different cardiac cycles to measure the a and b components to precisely define the beginning and end points; therefore, slight changes in heart rate between the time of measurement of a and b may be a source of error in these cases. In addition, it is important to consistently measure a and b components using a uniform leading and trailing edges technique, or another consistent method developed in one’s own laboratory with defined normal values. LVTX-MME results in this study were similar to those reported by Tham and Silverman.9 Between group variations in loading conditions, which were not analyzed in any of these studies, is another potential source of minor discrepancies between studies.15-17 We prefer the LVTX-TDI method because it requires imaging in only one view, thus, a and b components are measured in the same cardiac cycle in all cases. The advantage of TDI over PWD and MME is its capability to measure all time intervals directly in a single beat from a single view. This is highly advantageous, especially when measuring isovolumic relaxation or contraction times, which can be as short as 0.02 seconds in preterm or tachycardic infants. Our technicians rapidly mastered the fairly simple TDI technique and it has become a standard component of our clinical echocardiograms. Assuming the normal range for LVTX is the mean ⫾ 2SD, ie, up to a z score of ⫹2, the following LVTX values may be considered to be the upper limit of normal: LVTX-TDI less than 0.50, LVTX-PWD less than 0.50, and LVTX-MME less than 0.45.
REFERENCES 1. Dyer KL, Pauliks LB, Das B, Shandas R, Ivy D, Shaffer EM, et al. Use of myocardial performance index in pediatric patients with idiopathic pulmonary arterial hypertension. J Am Soc Echocardiogr 2006;19:21-7. 2. Dujardin KS, Tei C, Yeo TC, Hodge DO, Rossi A, Seward JB. Prognostic value of a Doppler index combining systolic and diastolic performance in idiopathic-dilated cardiomyopathy. Am J Cardiol 1998;82:1071-6. 3. Gaibazzi N, Petrucci N, Ziacchi V, del Garda D. Left ventricle myocardial performance index derived either by conventional method or mitral annulus tissue-Doppler: a comparison study in healthy subjects and subjects with heart failure. J Am Soc Echocardiogr 2005;18:1270-6. 4. Haque A, Otsuji Y, Yoshifuku S, Kumanohoso T, Zhang H, Kisanuki A, et al. Effects of valve dysfunction on Doppler Tei index. J Am Soc Echocardiogr 2002;15:877-83. 5. Harjai KJ, Scott L, Vivekananthan K, Nunez E, Edupuganti R. The Tei index: a new prognostic index for patients with symptomatic heart failure. J Am Soc Echocardiogr 2002;15: 864-8. 6. Bruch C, Schmermund A, Dagres N, Katz M, Bartel T, Erbel R. Severe aortic valve stenosis with preserved and reduced systolic
Journal of the American Society of Echocardiography Volume 19 Number 12
7.
8.
9.
10.
11.
12.
left ventricular function: diagnostic usefulness of the Tei index. J Am Soc Echocardiogr 2002;15:869-76. Williams RV, Ritter S, Tani LY, Pagotto LT, Minich LL. Quantitative assessment of ventricular function in children with single ventricles using the Doppler myocardial performance index. Am J Cardiol 2000;86:1106-10. Eidem BW, McMahon CJ, Cohen RR, Wu J, Finkelshteyn I, Kovalchin JP, et al. Impact of cardiac growth on Doppler tissue imaging velocities: a study in healthy children. J Am Soc Echocardiogr 2004;17:212-21. Tham EBC, Silverman NH. Measurement of the Tei index: a comparison of M-mode and pulse Doppler methods. J Am Soc Echocardiogr 2004;17:1259-65. Harada K, Tamura M, Toyono M, Oyama K, Takada G. Assessment of global left ventricular systolic function by tissue Doppler imaging. Am J Cardiol 2001;88:927-32. Tei C, Ling LH, Hodge DO, et al. New index of combined systolic and diastolic myocardial performance: a simple and reproducible measure of cardiac function–a study in normal and dilated cardiomyopathy. J Cardiol 1995;26:357-66. Tei C, Nishimura RA, Seward JB, Tajik AJ. Noninvasive Doppler-derived myocardial performance index: correlation
Cui and Roberson 1445
13.
14.
15.
16.
17.
with simultaneous measurements of cardiac catheterization measurements. J Am Soc Echocardiogr 1997;10:169-78. Quinones MA, Otto CM, Stoddard M, Waggoner A, Zoghbi WA. Recommendations for quantification of Doppler echocardiography: a report from the Doppler quantification task force of the nomenclature and standards committee of the American Society of Echocardiography. J Am Soc Echocardiogr 2002;15:167-84. Spencer KT, Kirkpatrick JN, Mor-Avi V, Decara JM, Lang RM. Age dependency of the Tei index of myocardial performance. J Am Soc Echocardiogr 2004;17:350-2. Moller JE, Poulsen SH, Egstrup K. Effect of preload alternations on a new Doppler echocardiographic index of combined systolic and diastolic performance. J Am Soc Echocardiogr 1999;135:1065-72. Levine SJ. Index of myocardial performance is afterload dependent in the normal and abnormal left ventricle. J Am Soc Echocardiogr 2005;18:342-50. Levine SJ. Effect of heart rate and preload on the index of myocardial performance in normal and abnormal left ventricle. J Am Soc Echocardiogr 2005;18:133-41.
The Tei index has been found to be useful for analyzing systolic and diastolic global ventricular function in a wide variety of congenital and acquired cardiac abnormalities. However, there are some discrepancies between reports as to the normal values for the Tei index obtained by the different echocardiographic techniques and by different investigators. We conducted a prospective study to determine the normal range of left ventricular Tei index (LVTX) values in a broad sample of children using tissue Doppler imaging, pulsed wave Doppler, and Mmode echocardiography. In all, 289 children with normal echocardiogram findings (age 1 day-18 years, body surface area 0.08-2.4 m2, heart rate 46-182/min) were studied. The LVTX was calculated by
The Tei index has become a widely used echocar-
diographic parameter for the assessment of global systolic and diastolic function in adults and children with a wide variety of congenital and acquired cardiac abnormalities.1-7 However, there are discrepancies between reports regarding the normal values in children, and variability in normal values depending on which echocardiographic modality is used to measure the Tei index.3,8-10 We undertook this prospective study to determine the normal values for the left ventricular (LV) Tei index (LVTX) measured from tissue Doppler imaging (TDI), pulsed wave Doppler (PWD), and M-mode echocardiography (MME). A cohort of children with normal echocardiogram results who spanned the range of ages, body sizes, and heart rates encountered in the pediatric population were studied. The normal values for each of the 3 methods were compared to determine any difference in the normal values depending on which technique was used. The effects of age, body surface area (BSA), and heart rate on the LVTX From the Heart Institute for Children, Hope Children’s Hospital. Reprint requests: David A. Roberson, MD, The Heart Institute for Children, Hope Children’s Hospital, 4440 W 95th St, Oak Lawn IL 60453 (E-mail: [email protected]). 0894-7317/$32.00 Copyright 2006 by the American Society of Echocardiography. doi:10.1016/j.echo.2006.06.006
1438
all 3 methods in each patient during a single echocardiographic examination. The normal LVTX values (mean ⴞ SD) for the 3 techniques were: LVTX-Doppler tissue imaging ⴝ 0.38 ⴞ 0.06; LVTX-pulsed wave Doppler ⴝ 0.36 ⴞ 0.07; and LVTX-M-mode echocardiography ⴝ 0.29 ⴞ 0.08. LVTX-Doppler tissue imaging and LVTX-pulsed wave Doppler values were only slightly but statistically significantly different (P < .05). LVTXM-mode echocardiography values were consistently and significantly less than those obtained by both of the other two methods (P < .01, respectively). The effects of age, body surface area, and heart rate were not clinically significant. These results are similar but not identical to those from prior studies. (J Am Soc Echocardiogr 2006;19:1438-1445.)
were analyzed. Lastly, our results were compared with those from previous studies.3,8-10 METHODS Study Design Tei index values obtained from TDI, PWD, and MME were each measured from the same normal echocardiogram result to determine the normal values for each technique. LVTX measurements were made at times of stable heart rate with the R-R interval varying by 40 milliseconds or less between measurements. Patients were grouped according to age as in prior studies.8 BSA was calculated and recorded in each case. The heart rate was calculated from the R-R interval at the time of LVTX measurements. Normal values for each technique were compared to determine whether the values obtained differed for the 3 techniques. Results for different age groups, BSAs, and heart rates were compared to determine any effect of these parameters on the LVTX obtained from all 3 methods. Patient Features The study group consisted of 289 pediatric patients with normal echocardiogram findings. Patient features are summarized in Table 1. The patient ages ranged from 1 day to 18 years. BSA spanned from 0.08 to 2.4 m2. The minimum heart rate was 46/min and the maximum heart rate was 182/min.
Journal of the American Society of Echocardiography Volume 19 Number 12
Cui and Roberson 1439
Table 1 Demographics in study patients n Male Age Range BSA, m2 Range HR, beats/min Range
<1 mo
1 mo-1 y
1-6 y
6-10 y
10-14 y
14-18 y
Total
46 23 7.3 ⫾ 7.3 d 1-29 d 0.187 ⫾ 0.06 0.08-0.40 139 ⫾ 14 113-182
35 16 3.0 ⫾ 2.3 mo 1-11 mo 0.28 ⫾ 0.07* 0.12-0.41 147 ⫾ 19* 90-181
50 28 2.8 ⫾ 1.2 y 1-5 y 0.61 ⫾ 0.11† 0.41-0.90 109 ⫾ 21† 71-162
45 20 7.3 ⫾ 1.2 y 6-9 y 0.96 ⫾ 0.18† 0.59-1.30 89 ⫾ 12† 55-115
44 22 11.7 ⫾ 1.0 y 10-13 y 1.45 ⫾ 0.29† 0.88-2.40 75 ⫾ 12† 46-107
69 40 15.5 ⫾ 1.1 y 14-18 y 1.74 ⫾ 0.21† 1.33-2.40 72 ⫾ 13 46-109
289 149 7.1 ⫾ 6.2 y 1 d-18 y 0.95 ⫾ 0.62 0.08-2.40 101 ⫾ 33 46-182
BSA, Body surface area; HR, heart rate. *P ⬍ .05; †P ⬍ .01 compared with preceeding age group. Data expressed as mean ⫾ SD.
Equipment Echocardiograms were performed by full-time clinical pediatric echocardiographic technicians and cardiology fellows or attending physicians. Three echocardiographic systems were used (Sonos 5500 and 7500, Philips Medical Systems, Andover, Mass; and Sequoia 512, Siemens Medical Solutions, Malvern, Pa). All images were stored digitally for further analysis and review (Xcelera system, Philips Medical Systems).
edge to the trailing edge of the left ventricular outflow tract (LVOT) PWD tracing (Figure 1, B). Both TDI and PWD recordings were made from the standard apical transducer position. LVTX-MME a component was measured from mitral valve closure to the subsequent mitral valve opening on the mitral valve MME tracing. The LVTX-MME b component was measured from aortic valve opening to aortic valve closure on the aortic valve MME tracing. MME recordings were obtained from the standard parasternal long axis view (Figure 1, C ).
Echocardiographic Studies Complete 2-dimensional, Doppler, color Doppler, and MMEs were performed on all patients. We include apical 4-chamber TDI of the mitral annulus in our standard clinical echocardiogram protocol. PWD of the mitral inflow and LV outflow, and MME of the mitral and aortic valves are also components of a standard complete echocardiogram. Echocardiograms were ordered for clinical purposes by the clinical attending physicians or house staff. For the purpose of this study, only echocardiograms with an official reading of completely normal study or completely normal study except for patent foramen ovale with a diameter of less than or equal to 2 mm and trivial left-to-right shunt were accepted for analysis. Tei Index Calculation The Tei index was calculated as previously described.3,8-14 Specific details of our measurement techniques were as follows. For all 3 methods the a component equals the sum of isovolumic contraction time plus ejection time plus isovolumic relaxation time. The b component is equal to the left ventricular ejection time. The LVTX is calculated as the difference of a minus b divided by b. The LVTX-TDI a component was measured from the trailing edge of the mitral annular A’ wave to the leading edge of the subsequent TDI mitral annular early diastolic (E’) wave. The LVTX-TDI b component was measured from the leading edge to the trailing edge of the TDI mitral annular systolic (S) wave (Figure 1, A). LVTX-PWD a component measurements were made from the trailing edge of the PWD late mitral A wave to the leading edge of the subsequent PWD early mitral E wave. The b component for the LVTX-PWD was measured from the leading
Statistical Analysis Continuous data are reported as the mean ⫾ SD. Discrete data are reported as frequencies and percentages. Oneway analysis of variance and analysis of variance for repeated measures analysis with post hoc Tukey test were used to determine significant differences of continuous data between groups. The effects of age, heart rate, and BSA on LVTX were evaluated by multiple correlation and linear regression analysis. Statistical significance was declared when the computed P value was less than .05. All statistical tests were 2-sided and all analyses were performed using software (SAS for Windows, SAS Inc, Cary, NC). Reproducibility Intraobserver variability was assessed in 22 participants by repeating the measurements on two occasions 5 days apart. To test the interobserver variability, the measurements were performed from digitally stored cycles by a second observer who was unaware of the prior results. The variability was calculated as the mean percentage error, derived as the difference between the two sets of measurements, divided by the overall mean of the observed values by the two measurements. Human Research This study complies with our institutional policies for human research and patient confidentiality.
1440 Cui and Roberson
Journal of the American Society of Echocardiography December 2006
Figure 1 A, Left ventricular (LV) Tei index (LVTX) calculation using tissue Doppler imaging (TDI). The a component (isovolumic contraction time ⫹ ejection time ⫹ isovolumic relaxation time) is measured from trailing edge of late diastolic TDI mitral annular A wave to leading edge of subsequent early diastolic TDI mitral annular E wave. The b component is measured from leading edge of systolic TDI mitral annular S wave to trailing edge of systolic TDI mitral annular S wave. LVTX formula is then calculated by difference of: a ⫺ b/b. B, LVTX calculation using tissue pulsed wave Doppler (PWD). The a component is measured from trailing edge of late diastolic transmitral PWD flow A wave to leading edge of subsequent early diastolic transmitral PWD flow E wave. The b component is measured from leading to trailing edge of left ventricular outflow systolic PWD tracing. LVTX formula is then calculated by difference of: a ⫺ b/b.
Journal of the American Society of Echocardiography Volume 19 Number 12
Cui and Roberson 1441
Figure 1 C, Left ventricular Tei index (LVTX) calculation using M-mode echocardiography (MME). The a component is from mitral valve closure to following mitral valve opening. The b component is from aortic valve opening to closing.
Figure 2 Left ventricular Tei index (LVTX) normal values for tissue Doppler imaging (TDI), pulsed wave Doppler (PWD), and M-mode echocardiography (MME) methods expressed as mean ⫾ SD.
RESULTS Normal LVTX values for the 3 techniques expressed as mean ⫾ SD (Figure 2) were: LVTX-TDI ⫽ 0.38 ⫾ 0.06; LVTX-PWD ⫽ 0.36 ⫾ 0.07; and LVTX-MME ⫽ 0.29 ⫾ 0.08. There was a small but statistically significant difference in LVTX-TDI versus LVTXPWD (P ⬍ .05). LVTX-MME was consistently and significantly less than LVTX-TDI and LVTX-PWD
(P ⬍ .01, respectively). Z scores for the 3 techniques are presented in Figure 3. Results for the normal values of LVTX-TDI, LVTXPWD, and LVTX-MME for various age groups are summarized in Table 2. There were some very small but statistically significant differences between some of the age groups for LVTX-TDI results. These included slightly lower LVTX-TDI normal values in age group 1 month to 1 year versus both age groups 10 to 14 years and 14 to 18 years (P ⬍ .05, respectively). There was also a slight difference in the age group 1 to 6 years LVTX-TDI being slightly less than the age group 14 to 18 years (P ⬍ .05). There was also a slight difference for LVTX-PWD results between patients age 1 month to 1 year versus both age groups 10 to 14 years and 14 to 18 years (P ⬍ .05, respectively). However, there was no significant association between LVTX by any of the 3 methods and age or heart rate after controlling for the effect of BSA. There was a statistical association between BSA and both LVTXTDI and LVTX-PWD (P ⬍ .01). R2 values of only 0.0579 and 0.0573, respectively, indicate only 5.8% of the variation in LVTX-TDI and 5.7% of the variation in LVTX-PWD can be explained by variation in BSA. Intraobserver and interobserver variability was low with measurement intraobserver variability of 3.0% ⫾ 2.0% and interobserver variability of 3.4% ⫾ 2.0%.
Journal of the American Society of Echocardiography December 2006
1442 Cui and Roberson
Figure 3 Left ventricular Tei index (LVTX) normal values for tissue Doppler imaging (TDI), pulsed wave Doppler (PWD), and M-mode echocardiography (MME) methods expressed as z scores. Table 2 Tei index for left ventricle derived from different methods TDI a
⬍1 mo
0.25 ⫾ 0.02 (0.24-0.26) 1 mo-1 y 0.25 ⫾ 0.03 (0.24-0.26) 1-6 y 0.34 ⫾ 0.04 (0.32-0.34) 6-10 y 0.36 ⫾ 0.03 (0.35-0.37) 10-14 y 0.39 ⫾ 0.02 (0.38-0.40) 14-18 y 0.39 ⫾ 0.03 (0.39-0.40) Overall 0.34 ⫾ 0.07 (0.33-0.35)
PWD
MME
b
Tei index
a
b
Tei index
a
b
Tei index
0.18 ⫾ 0.02 (0.18-0.19) 0.19 ⫾ 0.02 (0.18-0.20) 0.24 ⫾ 0.03 (0.23-0.25) 0.27 ⫾ 0.02 (0.26-0.27) 0.28 ⫾ 0.02 (0.27-0.28) 0.28 ⫾ 0.02 (0.28-0.29) 0.25 ⫾ 0.05 (0.24-0.25)
0.37 ⫾ 0.06 (0.35-0.40) 0.35 ⫾ 0.09‡§ (0.32-0.39) 0.36 ⫾ 0.04‡ (0.35-0.38) 0.37 ⫾ 0.04 (0.36-0.39) 0.40 ⫾ 0.07 (0.38-0.42) 0.40 ⫾ 0.05 (0.38-0.41) 0.38 ⫾ 0.06*† (0.37-0.39)
0.25 ⫾ 0.03 (0.24-0.25) 0.26 ⫾ 0.03 (0.24-0.25) 0.31 ⫾ 0.03 (0.30-0.32) 0.35 ⫾ 0.03 (0.34-0.36) 0.37 ⫾ 0.03 (0.36-0.38) 0.38 ⫾ 0.04 (0.37-0.39) 0.32 ⫾ 0.06 (0.31-0.33)
0.18 ⫾ 0.02 (0.18-0.19) 0.19 ⫾ 0.02 (0.18-0.19) 0.23 ⫾ 0.02 (0.22-0.23) 0.26 ⫾ 0.02 (0.25-0.26) 0.27 ⫾ 0.02 (0.26-0.28) 0.27 ⫾ 0.03 (0.27-0.28) 0.24 ⫾ 0.04 (0.23-0.24)
0.36 ⫾ 0.11 (0.32-0.39) 0.32 ⫾ 0.07‡§ (0.30-0.35) 0.36 ⫾ 0.04 (0.34-0.37) 0.35 ⫾ 0.05 (0.33-0.37) 0.38 ⫾ 0.07 (0.35-0.40) 0.39 ⫾ 0.07 (0.37-0.40) 0.36 ⫾ 0.07† (0.35-0.37)
0.24 ⫾ 0.02 (0.24-0.25) 0.25 ⫾ 0.02 (0.24-0.26) 0.30 ⫾ 0.03 (0.29-0.31) 0.33 ⫾ 0.03 (0.32-0.34) 0.36 ⫾ 0.03 (0.35-0.37) 0.37 ⫾ 0.04 (0.36-0.38) 0.32 ⫾ 0.06 (0.31-0.32)
0.19 ⫾ 0.02 (0.18-0.19) 0.20 ⫾ 0.03 (0.19-0.21) 0.24 ⫾ 0.03 (0.23-0.25) 0.26 ⫾ 0.02 (0.25-0.26) 0.28 ⫾ 0.02 (0.27-0.29) 0.29 ⫾ 0.03 (0.28-0.30) 0.25 ⫾ 0.05 (0.24-0.25)
0.30 ⫾ 0.09 (0.27-0.33) 0.27 ⫾ 0.09 (0.23-0.31) 0.28 ⫾ 0.05 (0.26-0.30) 0.28 ⫾ 0.07 (0.26-0.31) 0.29 ⫾ 0.07 (0.27-0.32) 0.29 ⫾ 0.09 (0.26-0.32) 0.29 ⫾ 0.08 (0.28-0.30)
a, isovolumic contraction time plus ejection time plus isovolumic relaxation time; b, ejection time; M, month; MME, M-mode echocardiography; PWD, pulse waved Doppler; TDI, tissue Doppler imaging; Y, years. *P ⫽ .0007 compared with pulsed Doppler-derived Tei index. †P ⬍ .0001 compared with M-mode-derived Tei index. ‡P ⬍ .05 compared with 14-18 y age group. §P ⬍ .05 compared with 10-14 y age group. Data expressed as mean ⫾ SD (95% confidence interval).
DISCUSSION These LVTX normal values can be applied to the entire spectrum of the pediatric population with no clinically significant effects of age, heart rate, and BSA throughout this group of patients. Although there were some slight statistical differences between LVTX results as a result of the effect of BSA, this difference of approximately 0.02 is so small that it seems negligible in the clinical setting. Once we controlled for the slight
effect of BSA, we did not find any effect of age on the LVTX in the pediatric population, in contrast to the findings of Spencer et al,14 who demonstrated some age-dependent changes of LVTX in adult patients. It was not surprising to find some differences between the LVTX values for the 3 methods, because they actually measure different time interval parameters for the a and b components of the LVTX. Typical differences in the timing and duration of a and b measurements are illustrated in the composite
Journal of the American Society of Echocardiography Volume 19 Number 12
Cui and Roberson 1443
Figure 4 Composite echocardiograms constructed from one echocardiogram to illustrate typical differences in timing and duration of a and b components of left ventricular (LV) Tei index (LVTX) for tissue Doppler imaging (TDI) (top), pulsed wave Doppler (PWD) (middle), and M-mode echocardiography (MME) (bottom). Events were timed with respect to electrocardiogram and TDI a and b components. A, Measurement of a component of LVTX by 3 methods. TDI a starts and ends later than PWD a and durations of a are similar for both. MME a starts at same time as TDI a but its duration is shorter than TDI a and PWD a. B, Measurement of b component of LVTX by 3 methods. TDI b starts and ends earlier than PWD b and durations of b are similar for both. MME b starts at same time as TDI b but its duration is longer than TDI b and PWD b. A, Mitral valve PWD A wave; A’, mitral annulus late diastolic myocardial velocity; AoV, aortic valve; E, mitral valve PWD E wave; E’, mitral annulus early diastolic myocardial velocity; LVOT, LV outflow tract; MV, mitral valve; R, R wave of QRS complex.
1444 Cui and Roberson
Figure 5 Between studies comparison of left ventricular Tei index (LVTX) normal values for tissue Doppler imaging (TDI), pulsed wave Doppler (PWD), and Mmode echocardiography (MME). Data are expressed as mean ⫾ SD. n, Number of patients in each study. Five studies compared are current study (Cui) and those of Eidem et al,8 Gaibazzi et al,3 Tham and Silverman,9 and Harada et al,10 respectively.
echocardiograms constructed for Figure 4. Figure 4, A, demonstrates that the TDI a begins after the PWD a begins and ends after the PWD a ends. The MME a begins at the same time as the TDI a but ends before both the TDI a or PWD a ends. Therefore, the a measurements are similar for TDI and PWD, but shorter for MME. Figure 4, B, illustrates that the TDI b wave begins slightly before the PWD b begins and that the TDI b ends slightly before the PWD b ends. The MME b begins at the same time as the TDI b but ends after both the TDI b and PWD b, resulting in a longer MME b versus both TDI b and PWD b. In other words, TDI and PWD a and b parameters have slightly different timing but similar duration, resulting in similar LVTX-TDI and LVTX-PWD. However, for MME, the a is shorter and the b is longer than the TDI and PWD parameters. This causes the LVTXMME to be smaller than the LVTX-TDI or LVTXPWD. There is extensive overlap between our results and those of other investigators (Figure 5). LVTXTDI results were similar to those obtained by Harada et al10 and Gaibazzi et al.3 The widest range of results between studies, albeit with significant overlap of the normal value range, was obtained by the PWD technique.3,8-10 A possible source of the minor discrepancies between studies may be difficulty in precisely measuring the termination point of mitral valve inflow (ie, the end of the mitral valve A wave)
Journal of the American Society of Echocardiography December 2006
by PWD because of low velocity noise around the baseline at this time in the cardiac cycle. In addition, LVTX-PWD measurement sometimes requires the use of different cardiac cycles to measure the a and b components to precisely define the beginning and end points; therefore, slight changes in heart rate between the time of measurement of a and b may be a source of error in these cases. In addition, it is important to consistently measure a and b components using a uniform leading and trailing edges technique, or another consistent method developed in one’s own laboratory with defined normal values. LVTX-MME results in this study were similar to those reported by Tham and Silverman.9 Between group variations in loading conditions, which were not analyzed in any of these studies, is another potential source of minor discrepancies between studies.15-17 We prefer the LVTX-TDI method because it requires imaging in only one view, thus, a and b components are measured in the same cardiac cycle in all cases. The advantage of TDI over PWD and MME is its capability to measure all time intervals directly in a single beat from a single view. This is highly advantageous, especially when measuring isovolumic relaxation or contraction times, which can be as short as 0.02 seconds in preterm or tachycardic infants. Our technicians rapidly mastered the fairly simple TDI technique and it has become a standard component of our clinical echocardiograms. Assuming the normal range for LVTX is the mean ⫾ 2SD, ie, up to a z score of ⫹2, the following LVTX values may be considered to be the upper limit of normal: LVTX-TDI less than 0.50, LVTX-PWD less than 0.50, and LVTX-MME less than 0.45.
REFERENCES 1. Dyer KL, Pauliks LB, Das B, Shandas R, Ivy D, Shaffer EM, et al. Use of myocardial performance index in pediatric patients with idiopathic pulmonary arterial hypertension. J Am Soc Echocardiogr 2006;19:21-7. 2. Dujardin KS, Tei C, Yeo TC, Hodge DO, Rossi A, Seward JB. Prognostic value of a Doppler index combining systolic and diastolic performance in idiopathic-dilated cardiomyopathy. Am J Cardiol 1998;82:1071-6. 3. Gaibazzi N, Petrucci N, Ziacchi V, del Garda D. Left ventricle myocardial performance index derived either by conventional method or mitral annulus tissue-Doppler: a comparison study in healthy subjects and subjects with heart failure. J Am Soc Echocardiogr 2005;18:1270-6. 4. Haque A, Otsuji Y, Yoshifuku S, Kumanohoso T, Zhang H, Kisanuki A, et al. Effects of valve dysfunction on Doppler Tei index. J Am Soc Echocardiogr 2002;15:877-83. 5. Harjai KJ, Scott L, Vivekananthan K, Nunez E, Edupuganti R. The Tei index: a new prognostic index for patients with symptomatic heart failure. J Am Soc Echocardiogr 2002;15: 864-8. 6. Bruch C, Schmermund A, Dagres N, Katz M, Bartel T, Erbel R. Severe aortic valve stenosis with preserved and reduced systolic
Journal of the American Society of Echocardiography Volume 19 Number 12
7.
8.
9.
10.
11.
12.
left ventricular function: diagnostic usefulness of the Tei index. J Am Soc Echocardiogr 2002;15:869-76. Williams RV, Ritter S, Tani LY, Pagotto LT, Minich LL. Quantitative assessment of ventricular function in children with single ventricles using the Doppler myocardial performance index. Am J Cardiol 2000;86:1106-10. Eidem BW, McMahon CJ, Cohen RR, Wu J, Finkelshteyn I, Kovalchin JP, et al. Impact of cardiac growth on Doppler tissue imaging velocities: a study in healthy children. J Am Soc Echocardiogr 2004;17:212-21. Tham EBC, Silverman NH. Measurement of the Tei index: a comparison of M-mode and pulse Doppler methods. J Am Soc Echocardiogr 2004;17:1259-65. Harada K, Tamura M, Toyono M, Oyama K, Takada G. Assessment of global left ventricular systolic function by tissue Doppler imaging. Am J Cardiol 2001;88:927-32. Tei C, Ling LH, Hodge DO, et al. New index of combined systolic and diastolic myocardial performance: a simple and reproducible measure of cardiac function–a study in normal and dilated cardiomyopathy. J Cardiol 1995;26:357-66. Tei C, Nishimura RA, Seward JB, Tajik AJ. Noninvasive Doppler-derived myocardial performance index: correlation
Cui and Roberson 1445
13.
14.
15.
16.
17.
with simultaneous measurements of cardiac catheterization measurements. J Am Soc Echocardiogr 1997;10:169-78. Quinones MA, Otto CM, Stoddard M, Waggoner A, Zoghbi WA. Recommendations for quantification of Doppler echocardiography: a report from the Doppler quantification task force of the nomenclature and standards committee of the American Society of Echocardiography. J Am Soc Echocardiogr 2002;15:167-84. Spencer KT, Kirkpatrick JN, Mor-Avi V, Decara JM, Lang RM. Age dependency of the Tei index of myocardial performance. J Am Soc Echocardiogr 2004;17:350-2. Moller JE, Poulsen SH, Egstrup K. Effect of preload alternations on a new Doppler echocardiographic index of combined systolic and diastolic performance. J Am Soc Echocardiogr 1999;135:1065-72. Levine SJ. Index of myocardial performance is afterload dependent in the normal and abnormal left ventricle. J Am Soc Echocardiogr 2005;18:342-50. Levine SJ. Effect of heart rate and preload on the index of myocardial performance in normal and abnormal left ventricle. J Am Soc Echocardiogr 2005;18:133-41.