European Journal of Echocardiography Advance Access originally published online on March 11, 2008
European Journal of Echocardiography 2008 9(4):542-546; doi:10.1093/ejechocard/jen114
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Longitudinal plane colour tissue-Doppler myocardial velocities and their association with left ventricular length, volume, and mass in humans
1 Centre for Food, Physical Activity, and Obesity Research, University of Teesside, Middlesbrough, UK
2 Centre for Sport Medicine and Human Performance, Brunel University, Uxbridge, London, UK
3 School of Healthcare, Leeds University, Leeds, UK
4 Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, 15–21 Webster Street, Liverpool L3 2ET, UK
Received 16 May 2007; accepted after revision 30 September 2007; online publish-ahead-of-print 11 March 2008.
* Corresponding author. Tel: +44 151 231 4088; fax: +44 151 231 4353. E-mail address: k.george{at}ljmu.ac.uk
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Aims: We investigated the relationships between colour tissue-Doppler parameters of left ventricular (LV) function and indices of LV morphology.
Methods and results: LV length, end-diastolic volume, and mass were assessed in 40 healthy adult subjects. Further, colour tissue-Doppler scans assessed peak myocardial velocities during systole (S') and early diastole (E') as well as acceleration during isovolumic contraction (IVCa) at the mitral annulus. Non-linear allometric relationships (Y = aXb) were calculated to provide size exponents (b), with 95% confidence intervals, for tissue-Doppler variables (Y) and LV morphology parameters (X). The b exponents for LV length with Peaks S' and E' were not substantially different from unity (b = 0.87 and 0.95, respectively, P > 0.05). Peak E' was also associated with LV volume (b = 0.39, r = 0.53, P < 0.05). IVCa was not related to any LV morphology parameter.
Conclusion: Peak S' or E' divided by LV length was confirmed as a valid size-independent scaling index. Conversely, IVCa is seemingly size independent.
Keywords: Scaling; Normalisation; Allometry; Athlete's heart
Tissue-Doppler imaging (TDI) is rapidly gaining recognition as an important clinical tool in the assessment of ventricular function1 largely due to the fact that myocardial tissue velocities can be measured with high reproducibility and allows interrogation of both segmental and global left ventricular (LV) function.2–4 The application of TDI has included the determination of the global and segmental impact of athletic training upon LV and right ventricular (RV) performance5–8 as well as the differentiation of pathologic disease states from physiologic LV adaptation.9,10
The impact of LV morphology upon the peak colour tissue velocities reported during systole and early diastole has received limited attention and remains controversial. Such issues are central to our understanding of the need to scale colour tissue-Doppler data to facilitate intra- and inter-individual comparisons. While there is evidence that global indices of systolic function should be scaled allometrically for body size,11,12 the necessity to scale tissue-Doppler data for individual differences in LV morphology has received scant attention.8,13 This issue is pertinent to ‘athletic heart’ research because of the changes in LV morphology induced by chronic training.14,15 Although not clearly communicated, it is assumed that Pelà et al.8 ‘normalized’ pulsed tissue-Doppler myocardial velocities at the mitral annulus in athletes and controls by calculating ‘fractional’ tissue velocities that were simply divided by the short- or long-axis dimension of the LV. This removed any impact of training on pulsed tissue velocities. Scaling by this per ratio standards approach is, however, valid only if the relationship between tissue-Doppler velocities and the chosen parameter of ventricular morphology is linear and meets Tanner’s special circumstance.16 This empirical relationship was not confirmed by Pelà et al.8 and a recent comparative physiology study by Popovi
et al.13 suggested that the relationships between LV end-diastolic volume and/or mass with mitral annular colour tissue velocities were not linear and followed an allometric model with slope components of 0.096 ± 0.012–0.100 ± 0.013. Such empirical data prompt the re-interpretation of any conclusions based on simple ratio normalization methods.
One component of the tissue-Doppler trace that has received recent attention is myocardial performance associated with the isovolumic phases of the cardiac cycle. Shimizu et al.17 suggested that isovolumic acceleration was more sensitive for detecting LV dysfunction than any ejection phase tissue-Doppler index and that isovolumic acceleration was significantly correlated with the invasive assessment of dP/dtmax. Vogel et al.18 validated tissue-Doppler assessment of LV isovolumic acceleration (IVCa) in an animal model and Hashimoto et al.19 concluded that IVCa was a preload-independent marker of LV function, although this has been contradicted.20 To the best of the authors’ knowledge, the relationship of the IVCa with ventricular morphology has not been studied previously.
Therefore, the aim of this study was to provide empirical evidence, in humans, of the nature of relationships between colour tissue-Doppler myocardial velocities measured at the mitral annulus and various parameters of LV morphology. Specifically, we wanted to provide unique data for tissue acceleration obtained during isovolumic contraction. We hypothesized that these relationships would be statistically significant and substantial, but non-linear.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Subjects
We recruited 40 healthy adult subjects (35 males and 5 females). Basic subject details are provided in Table 1. The subjects were all physically active and trained for endurance competitions. The group represented a broad age range (24–59 years) and a diverse level of athletic performance (personal best marathon time 2:45–4:30 h:min). They were free from known cardiovascular disease and any early family history of cardiovascular disease, and were not currently taking any form of prescribed medication. After familiarization and a full explanation of data collection procedures, all subjects completed a written informed consent declaration. The study was granted by local Ethics Committee approval.
|
Design and protocols
All subjects undertook their evaluation in the week prior to a race performance. Upon arrival at the testing area, subjects were initially assessed for body mass (BM), in running with shorts and vest, on standard portable scales (Model A3JJT1K, Hansen, UK) and height via stadiometry. Subjects then lay supine for 5 min before duplicate brachial artery systolic and diastolic blood pressures were assessed by standard auscultation. Also, at this time, a resting heart rate was recorded from ECG within the echocardiography system.
Two-dimensional, M-mode, and colour TDI echocardiographic scans were performed using a commercially available ultrasound system (Vivid7, GE, Hertfordshire, UK) with a 2.5–4 MHz phased array transducer. All image acquisitions were made by a single, experienced sonographer with the subject lying in the left lateral decubitas position.
Using apical two- and four-chamber views, longitudinal colour tissue velocities at the mitral annulus of the lateral, septal, anterior, inferior, and posterior LV walls were recorded in systole (S') and early diastole (E') as well as tissue acceleration during isovolumic contraction. The sample volume was set at 6 x 6 mm and the image angle was adjusted to ensure parallel alignment between wall motion and the ultrasound beam. Gain, filter, pulse repetition frequency (PRF), sector size, and depth were adjusted to optimize colour saturation and obtain the highest frame rates (>100 fps). Two-dimensional echocardiographic recordings were obtained from apical two- and four-chamber views. All system settings were optimized to produce best signal-to-noise ratio and provide optimal endocardial definition. LV length, volume, and mass were determined according to standard measurement guidelines.21 Specifically, LV length was assessed at end-diastole as the longest dimension from two- and four-chamber images. LV length was also assessed at the point of peak tissue velocity during systole. LV end-diastolic volume was calculated using the Biplane method. LV mass was calculated from M-mode traces at the level of the mitral valve (LV mass = 0.8 x (1.04[(LVIDd + SWTd + PWTd)3—(LVIDd3)]) +0.6 g). All images were stored digitally for later analysis.
All images were analysed offline using dedicated software (Echopac PC, version 5.0.1, GE Medical, Horten, Norway) by a single, experienced sonographer who was blinded to subject identity. A minimum of three consecutive cardiac cycles were analysed for all echocardiographic indices. Myocardial velocity profiles were averaged over the five segments to provide a global value for colour tissue-Doppler velocities and acceleration. Peak IVCa was calculated by measuring the steepest slope of positive deflection during the isovolumic phase of systole. This was confirmed by establishing aortic valve opening and closure using curved anatomical M-mode.
Data analysis and statistics
The tissue-Doppler variables used in all analyses were the given by the mean of five measurement sites. Relationships between Peaks S', E', IVCa (Y) and LV length, volume, and mass as scaling denominators (X) were examined using the general allometric model Y = aXb. To avoid the potential bias associated with fitting a log-transformed linear model,22,23 working in the arithmetic space defined by the original, raw X, and Y variables we solved all allometric regression model parameters by an iterative, non-linear protocol using the Levenberg–Marquardt algorithm.24 In this procedure, small successive corrections to parameter estimates are made until a global solution converges. Size exponents (b) together with their 95% CIs were calculated and presented. Where b exponents were not substantially different from unity, the validity of ratio standards scaling (Y/X) was further examined using Tanner’s15 ‘special circumstance’ calculation. In this procedure, the correlation coefficient (r[x, y]) was compared with the ratio of the coefficients of variation (CV) for the same two variables (CVx/CVy). The ratio scaling is valid if r[x, y] is not substantially different from CVx/CVy. Bivariate correlations between Y/Xb (or Y/X where b is not substantially different from unity) and X were examined to confirm that any derived scaled index is size independent. The correlation between the index and the size variable should not differ substantially from zero.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Data for all pertinent parameters of LV morphology are reported in Table 1. Data for LV mass were at the upper end of the normal range probably due to the trained status of the subjects. Descriptive cohort data for colour tissue-Doppler myocardial velocities and IVCa are presented in Table 2. There are some small inter-site differences, but all data are within or just above normal ranges,25 again likely reflective of training status.
|
Data for b exponents and their 95% confidence limits are presented in Table 3. For both Peak S' and Peak E', the b exponents for scaling by LV length are not significantly different from unity. Regarding Tanner’s special circumstance, to test the validity of ratio scaling by LV length, for Peak S', the CVx/CVy = 0.38 (95% CI, 0.28–0.53) and r(x, y) = 0.33 (95% CI, 0.02–0.58). For Peak E', CVx/CVy = 0.38 (95% CI, 0.28–0.52) and r(x, y) = 0.35 (95% CI, 0.04–0.60). Hence, Tanner’s special circumstance is satisfied, as CVx/CVy is not substantially different from r(x, y). Correlations between the scaled ratio indices—Peak S'/LV length and Peak E'/LV length—and LV length were –0.08 (95% CI, –0.38 to 0.24) and –0.04 (95% CI, –0.35 to 0.27), respectively. These correlations are not significantly or substantially different from zero, indicating that the indices are size independent.
|
The b exponent for scaling Peak S' by LV length at peak systolic velocity was also not significantly different from unity. However, there was a larger discrepancy between CVx/CVy (0.57; 95% CI, 0.42–0.78) and r(x, y) (0.41; 95% CI, 0.11–0.64). Furthermore, the correlation between the ratio scaled index (Peak S'/LV length at peak systolic velocity) and LV length at peak systolic velocity was –0.19 (95% CI, –0.47 to 0.13), suggesting that this ratio standard tends to over-correct for the influence of this size variable. The power function ratio Peak S'/LV length at peak systolic velocity0.72, however, was size independent (r = –0.01; 95% CI, –0.32 to 0.30 between the power function ratio and the size variable).
Peak E' was also associated with LV volume, with an exponent (0.39) not substantially different from 1/3 (analogous to the length dimension assuming dimensional consistency). None of the tissue-Doppler variables were significantly related to LV mass. Peak IVCa was not significantly related to any scaling denominator. For the significant size exponents reported, non-linear correlation coefficients for the allometric models ranged from 0.33 to 0.53. Correlation coefficients may be converted to standardized effect sizes (Cohen’s d, in standard deviation units) using Friedman’s26 formula: 
. Hence, the range of correlation coefficients above is equivalent to effect sizes of 0.7–1.25 standard deviations—moderate to large effects on Cohen’s27 scale of magnitudes.
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The key finding of this study was that both S' and E', derived as the mean of five segmental measurements, were proportional to LV length in a cohort of physically active adults. The derived relationships support the importance of scaling (or normalization) of colour tissue-Doppler myocardial velocities in future research.
The finding that both S' and E' were proportional to LV length provides empirical support for the scaling procedures adopted by Pela et al.8 in a between-group comparison of athletes and sedentary controls. The proportionality is likely due to the nature of the length–tension relationship observed in human hearts28 that has been known for many years. The nature of the relationships reported between LV length and S' as well as E' allows the relatively simple ratio standard scaling process (Y/X) to be adopted, which is important due to its ‘user-friendly’ nature. The current data, and that of Pela et al.,8 would suggest that any athlete-control comparison of systolic or early diastolic tissue velocities that did not adopt empirically derived or supported scaling5–7 may have produced inappropriate conclusions. Such studies or populations may be revisited in future research. The exponents derived in the current study differ from those reported by Popovi et al.13 The smaller exponents reported by Popovi et al.13 are likely due to the use of multiple animal cohorts of vastly differing body sizes. While such scaling data are important, theoretically their application to single-species data analysis is potentially limited.
Scaling E' by LV volume also conformed to the concepts of geometric similarity as the b exponent was not significantly different from 1/3, i.e. proportional to length assuming dimensional consistency. As further confirmation, in this small sample, the b exponent derived from scaling LV volume by LV length is not significantly different from 3, i.e. the cube of length (95% CI for b exponent is 1.40–3.1). The use of LV volume, however, is slightly more complex mathematically, and the LV volume is more time consuming to assess than LV length. Likewise, LV length at peak tissue velocity was associated with S', but scaling to unity did not provide a size-independent index. Thus, LV volume or LV length at peak tissue velocity is not advocated to replace LV length when scaling S' or E' in this cohort. Finally, LV mass would seem to be a poor choice as a scaling index, likely due to the geometric complexity/assumptions involved in its calculation and/or the diverse patterns of LV wall and chamber geometry in this cohort.
Indirect evidence that peak colour tissue-Doppler velocities are preload dependent is provided by the linear relationships between S' or E' and LV length, as well as the non-linear relationship between E' and LV volume. The relative preload independence of tissue-Doppler velocities has been reported as a strength of such data.2 Preload dependence/independence continues to be hotly debated and remains a topic for further study. Interestingly, the lack of association between IVCa and LV morphology may suggest that this parameter is preload independent. This has been suggested before,19 but is also contentious.20 Currently, we would support the reporting of IVCa without any consideration for normalization by LV morphology.
Finally, it is important to evaluate the clinical relevance of the current data while bearing in mind caveats of allometric scaling and some limitations within the current dataset. The following comments also serve as a guide for future research directions. The current data clearly demonstrate the impact of LV morphology, and specifically LV length, upon basal LV Peak S' and Peak E' longitudinal myocardial velocities. If meaningful clinical and scientific comparisons of individuals or groups, with disparate LV morphology, are to be derived, then scaling for LV length would seem appropriate. Without scaling, the ability to provide clear interpretation is limited. While conceptually we support this, the application of our data must be circumspect. Allometry is sample specific and our current sample size and heterogeneity is limited. For balance, we provide some caution to the rapid adoption of any scaling practice.
The current data do prompt clinicians and scientists assessing and interpreting myocardial tissue velocity data to consider some specific issues. The adoption of any scaling practice is best served by empirical guidance from within the sample itself. In other words, if you want to scale for any index of LV morphology, assess the relationship first in your own data and this will provide the best guide to generating size-independent data. Using theoretically supported scaling practice (such as adopting scaling by LV length, as we have demonstrated) may be useful, but care must be taken to clearly denote if the scaled variable is indeed size independent (via simple correlation analysis). Further analyses that support or refute the current findings in a range of different populations are required. Only in this way will the full impact be elucidated on clinical decision making. Initially, it may be pertinent to simply increase the sample size, assess myocardial velocities in different segments and planes, and then interrogate the importance of LV morphology. This would likely serve to offset concerns with the broad confidence limits for all b exponents in our data. The broad limits likely reflect the relatively small sample size and the limited range of dependent and independent variables and suggest that other variables account for a significant proportion of the variability in colour tissue-Doppler data. Future research may then move onto the assessment of relationships between RV tissue velocities and parameters of RV morphology. We did not report data for isovolumic deceleration rate at the onset of diastole due to the lack of clear guidelines for assessment. Collection of this data may also provide unique insights. As the assessment of segmental LV function has developed variables such as strain and strain rates are being widely reported.29 As measurements of deformation and the rate of deformation we currently do not know whether gross LV morphology mediates strain and strain rate data and this should be investigated.
In conclusion, this paper provides empirical support for the necessity to scale longitudinal colour tissue-Doppler velocities when making between-subject comparisons where participants have different LV morphology. Specifically, the use of LV length as a per-ratio standard scaling variable for S' and E' is supported in the current cohort, but this requires further research and confirmation in different samples. The lack of association between IVCa and LV morphology may be interpreted to support claims of preload independence for this specific variable. While the concept of scaling such data can no longer be simply ignored, further research in this area is required.
|
Acknowledgements |
|---|
We are grateful to our subjects and the all those at Cardiac Risk in the Young who helped facilitate this work. We thank GE, UK and SA Ltd. for providing technical assistance.
Conflict of interest: none declared.
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
- Cheuk-Man Y, Sanderson JE, Marwick TH, Oh JK. Tissue Doppler imaging: a new prognosticator for cardiovascular diseases. J Am Coll Cardiol (2007) 49:1903–14.
[Abstract/Free Full Text] - Andersen NH, Terkelsen CJ, Sloth E, Poulsen SH. Influence of preload alterations on parameters of systolic left ventricular long-axis function: a Doppler tissue study. J Am Soc Echocardiogr (2004) 17:941–7.[CrossRef][Web of Science][Medline]
- Nikitin NP, Witte KK. Application of tissue Doppler imaging in cardiology. Cardiology (2004) 101:170–84.[CrossRef][Web of Science][Medline]
- Nikitin NP, Loh PH, de Silva R, Ghosh J, Khaleva OY, Goode K, et al. Prognostic value of systolic mitral annular velocity measured with Doppler tissue imaging in patients with chronic heart failure caused by left ventricular systolic dysfunction. Heart (2006) 92:775–9.
[Abstract/Free Full Text] - Schmidt-Trucksäss A, Schmidt A, Häussler C, Huber G, Huonker M, Keul J. Left ventricular wall motion during diastolic filling in endurance-trained athletes. Med Sci Sports Exerc (2001) 32:189–95.
- Caso P, Galderisi M, D’Andrea A, Di Maggio D, De Simone L, Martinello AR, et al. Analysis by pulsed-Doppler tissue imaging of ventricular interaction in long distance competitive swimmers. Am J Cardiol (2002) 90:193–7.[CrossRef][Web of Science][Medline]
- Vinereanu N, Florescu N, Sculthorpe N, Tweddel AC, Stephens MR, Fraser AG. Left ventricular long axis diastolic function is augmented in the hearts of endurance-trained compared with strength-trained athletes. Clin Sci (2002) 103:249–57.[Web of Science][Medline]
- Pelà G, Brushci G, Montagna L, Manara M, Manca C. Left and right ventricular adaptation assessed by Doppler tissue echocardiography in athletes. J Am Soc Echocardiogr (2004) 17:205–11.[CrossRef][Web of Science][Medline]
- Palka P, Lange A, Fleming AD, Donnelly JE, Dutka DP, Starkey IR, et al. Differences in myocardial velocity gradient measured throughout the cardiac cycle in patients with hypertrophic cardiomyopathy, athletes and patients with left ventricular hypertrophy due to hypertension. J Am Coll Cardiol (1997) 30:760–8.[Abstract]
- Vinereanu N, Florescu N, Sculthorpe N, Tweddel AC, Stephens MR, Fraser AG. Differentiation between pathological and physiological left ventricular hypertrophy by tissue Doppler assessment of long-axis function in patients with hypertrophic cardiomyopathy or hypertension and in athletes. Am J Cardiol (2001) 87:1226–30.[CrossRef][Web of Science][Medline]
- de Simone G, Devereux RB, Daniels SR, Mureddu G, Roman MJ, Kimball TR, et al. Stroke volume and cardiac output in normotensive children and adults. Assessment of relations with body size and impact of overweight. Circulation (1997) 95:1837–43.
[Abstract/Free Full Text] - Turley KR, Stanforth PR, Rankinen T, Bouchard C, Leon AS, Rao DC, et al. Scaling submaximal exercise cardiac output and stroke volume: The HERITAGE Family Study. Int J Sports Med (2006) 27:993–9.[CrossRef][Web of Science][Medline]
- Popovi ZB, Ping Sun J, Yamada H, Drinko J, Mauer K, Greenberg NL, et al. Differences in left ventricular long-axis function from mice to humans follow allometric scaling to ventricular size. J Physiol (2005) 568:255–65.
[Abstract/Free Full Text] - Pluim BM, Zwinderman AH, van der Laarse A, van der Wall EE. The athlete’s heart: a meta-analysis of cardiac structure and function. Circulation (1999) 100:1038–44.
- Whyte GP, George K, Nevill A, Shave R, Sharma S, McKenna WJ. Left ventricular morphology and function in female athletes: a meta-analysis. Int J Sports Med (2004) 25:380–3.[CrossRef][Web of Science][Medline]
- Tanner JM. Fallacy of per-weight and per-surface area standards, and their relation to spurious correlation. J Appl Physiol (1949) 2:1–5.
[Free Full Text] - Shimizu M, Nii M, Konstantinov IE, Li J, Redington AN. Isovolumic but not ejection phase Doppler tissue indices detect left ventricular dysfunction caused by coronary stenosis. J Am Soc Echocardiogr (2005) 18:1241–6.[CrossRef][Web of Science][Medline]
- Vogel M, Cheung MEH, Li J, Kristiansen SB, Schmidt MR, White PA, et al. Noninvasive assessment of left ventricular force frequency relationships using tissue Doppler-derived isovolumic acceleration: validation in an animal model. Circulation (2003) 107:1647–52.
[Abstract/Free Full Text] - Hashimoto I, Li X-K, Hejmadi-Bhat A, Jones M, Sahn DJ. Quantitative assessment of regional peak myocardial acceleration during isovolumic contraction and relaxation time by tissue Doppler imaging. Heart (2005) 91:811–6.
[Abstract/Free Full Text] - Lyseggen E, Rabben SI, Skulstad H, Urheim S, Risoe C, Smiseth OA. Myocardial acceleration during isovolumic contraction: relationship to contractility. Circulation (2005) 111:1362–9.
[Abstract/Free Full Text] - Lang RM, Bierig M, Devereux RB, Flaschkampf FA, Foster E, Pellikka PA, et al. Recommendations for chamber quantitation: a report from the American Society of Echocardiogrpahy’s guidelines and standards committee and the chamber quantitation writing group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr (2005) 18:1440–63.[CrossRef][Web of Science][Medline]
- Albrecht GH, Gelvin BR, Hartman SE. Ratios as a size adjustment in morphometrics. Am J Physical Anthrop (1993) 91:441–68.[CrossRef]
- Zar JH. Calculation and miscalculation of the allometric equation as a model in biological data. Bioscience (1968) 18:1118–20.[CrossRef][Web of Science]
- Glass NR. Discussion of calculation of power function with special reference to respiratory metabolism in fish. J Fisheries Res Board Can (1969) 26:2643–50.
- Nikitin NP, Witte KK, Thackrey SDR, de Silva R, Clark AL, Cleland JGF. Longitudinal ventricular function: normal values of atrioventricular annular and myocardial velocities measured with quantitative two-dimensional colour Doppler tissue imaging. J Am Soc Echocardiogr (2003) 16:906–21.[CrossRef][Web of Science][Medline]
- Friedman H. Magnitude of experimental effect and a table for its rapid estimation. Psych Bull (1968) 70:245–51.[CrossRef][Web of Science]
- Cohen J. Statistical power analysis for the behavioral sciences. (1988) 2nd ed. New Jersey: Lawrence Erlbaum.
- Sonnenblick E. Force-velocity relations in mammalian heart muscle. Am J Physiol (1962) 202:931–9.
[Abstract/Free Full Text] - Marwick TH. Measurement of strain and strain rate by echocardiography: ready for prime time? J Am Coll Cardiol (2006) 47:1313–27.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Oxborough, A. M. Batterham, R. Shave, N. Artis, K. M. Birch, G. Whyte, P. N. Ainslie, and K. P. George Interpretation of two-dimensional and tissue Doppler-derived strain ({varepsilon}) and strain rate data: is there a need to normalize for individual variability in left ventricular morphology? Eur J Echocardiogr, July 1, 2009; 10(5): 677 - 682. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||