Article Text

Download PDFPDF

Alterations in myocardial stiffness in elite athletes assessed by a new Doppler index
  1. G J King,
  2. R T Murphy,
  3. I Almuntaser,
  4. K Bennett,
  5. E Ho,
  6. A S Brown
  1. Cardiology Department, St James’s Hospital, Dublin, Ireland
  1. Dr G J King, Cardiology Department, St James’s Hospital, James’s Street, Dublin 8, Ireland; gking{at}


Background: In elite athletes left ventricular (LV) morphological changes are predicted to alter passive pressure/volume characteristics by reducing myocardial stiffness and increasing compliance.

Objective: To investigate the utility of a new Doppler tissue index based on the pressure volume relation ((E/Ea)/LVEDD), which provides a measure of myocardial stiffness, and to assess its usefulness in detecting cardiac adaptation in elite rowers.

Methods: Thirty-six international rowers who had trained intensively and a control group of 30 sedentary subjects, matched for age and sex, were enrolled in the study. LV septal and posterior wall thickness, mass, chamber size, transmitral Doppler peak early (E) and late (A) diastolic filling velocities and isovolumic relaxation times were measured. Early diastolic myocardial velocities (Ea) were averaged from four sites at the mitral annulus; diastolic stiffness was assessed with the use of three indices E, Ea and the LV end-diastolic diameter in diastole (LVEDD). The ratio, (E/Ea)/LVEDD, provides a new index of diastolic stiffness. Rowers were further divided into two groups based on the presence or absence of left ventricular hypertrophy (LVH) ⩽12 mm and >12 mm.

Results: No significant difference in Ea was found between the two groups, but there was a difference in the stiffness index, which remained after adjustment for body surface area and heart rate (controls 1.48 (0.3) vs athletes 1.17 (0.34), p = 0.016). No difference in stiffness index was found between the groups with LVH ⩽12 mm and >12 mm (1.11 (0.32) vs 1.17 (0.34), respectively, p = 0.68)

Conclusions: Intense training reduces myocardial stiffness despite the development of LVH.

View Full Text

Statistics from

Doppler tissue imaging (DTI) is an emerging modality in the assessment of athletic cardiac physiology and in sports medicine.1 The extent of left ventricular (LV) morphological changes varies with different training programmes in athletes. These changes are predicted to alter the passive pressure–volume characteristics of the left ventricle by reducing myocardial stiffness and increasing compliance.2 3 Few studies have specifically examined adaptations of the athletic heart using DTI.1 We therefore investigated the utility of a new Doppler tissue index based on the pressure/volume relationship, (E/Ea)/LVEDD, derived from transmitral peak E velocity, DTI peak Ea velocity and left ventricular end-diastolic chamber dimensions (LVEDD), which provides a measure of myocardial stiffness (the inverse of compliance), and assessed its usefulness in detecting cardiac adaptation in elite rowers.4


Patient selection

Between March 2004 and September 2005, 36 athletes were identified. They were international rowers, who had trained intensively for between 15 and 20 h a week for more than 5 years. None had a family history of hypertrophic cardiomyopathy or premature sudden death age <40 years and all were normotensive. A control group of 30 sedentary but otherwise normal subjects matched for age and sex were recruited from hospital personnel. Participants were enrolled after giving informed consent. Subjects were excluded if they had an abnormal rhythm, known coronary artery disease or significant valvular disease, hypertension or if echocardiographic images were technically inadequate for comprehensive analysis. The local hospital research committee approved this study.

Echocardiographic examination

Echocardiographic studies were performed using a Philips Sonos 5500-ultrasound system (Andover, Massachusetts, USA) equipped with DTI capabilities. M-mode and two-dimensional measurements were made according to the recommendations of the American Society of Echocardiography.5 Sepal and posterior wall thicknesses and LV mass and chamber sizes were determined from two-dimensional and M-mode images. The M-mode-derived anteroposterior linear dimension of the left atrium (LA) was obtained from the parasternal long-axis view. Peak early (E) and late (A) transmitral filling velocities and the isovolumic relaxation time (IVRT) were measured using conventional Doppler ultrasound. From apical windows a 10 mm sample volume was placed at four sites in the mitral annulus and spectral DTI was recorded on paper at 50 mm/s sweep speed. From these four sites, peak systolic (Sa), early Ea and late Aa diastolic velocities were measured and averaged.

Derived index

Diastolic stiffness was assessed with the use of three indices E, Ea and the LVEDD. The ratio E/Ea represents an index of LV filling pressure6 and LVEDD was used as an index of ventricular volume. The ratio of these two parameters, (E/Ea)/LVEDD, provides a new index of diastolic stiffness and this was measured for each subject in the study.

Statistical methods

Results are presented as mean (SD) where the data are normally distributed and as median (interquartile range (IQR)) where the data are non-normally distributed. Differences between control and athlete groups were determined using either the independent Student t test (two-group comparison) or analysis of variance (three-group comparison) for normally distributed data and using the non-parametric Wilcoxon rank sum test otherwise. Multiple regression analysis was used to adjust for body surface area and heart rate when examining differences between groups for LVEDD and diastolic stiffness index. A p value <0.05 was considered significant. All statistical analyses were performed using the JMP statistical analysis package (SAS).


Table 1 presents the clinical characteristics of the study group. There were significant differences between subjects and controls for heart rate (p<0.001) and body surface area (BSA; p<0.001).

Table 1 Baseline characteristics of the study group

Table 2 shows the echocardiographic data for the normally distributed variables ejection fraction (EF%), mean Ea velocities, LVEDD, transmitral E velocity and E/A ratio and the new diastolic stiffness index. Controlling for differences in heart rate and BSA, LVEDD was significantly greater, and the new diastolic stiffness index significantly lower, in athletes than in controls.

Table 2 Analysis of normally distributed echocardiographic variables by control and athlete groups

Table 3 shows a comparison of the non-normally distributed septal thickness, IVRT, E/Ea and LA size between the two groups. Septal wall thickness was significantly greater in athletes.

Table 3 Non-normally distributed echocardiographic variables of the study group

Table 4 shows the differences in Ea velocities and the new diastolic stiffness index in athletes according to the absence or presence of LVH in this group. Both diastolic function and LV compliance, as represented by the new stiffness index, did not differ between the two subgroups.

Table 4 Echocardiographic variables of the study group according to the absence or presence of left ventricular hypertrophy in the athlete group


Given the recent interest in reducing the risk of sudden cardiac death, there is great emphasis on the early detection of cardiac abnormalities in modern elite athletes.7 8 Prolonged intense training is characterised by left ventricular hypertrophy (LVH) and chamber dilatation, so it is important to distinguish physiological changes from pathological abnormalities. Doppler tissue indices may be useful in this regard. In our study of elite rowers an increase in LVEDD was present, with a modest parallel increase in septal and posterior wall thickness. As in previous studies, we found that LV systolic function, expressed as EF (table 2), was similar to that found in untrained subjects.9 At this level of exercise the end-diastolic and end-systolic volumes increase in parallel, increasing the stroke volume, and so the EF remains the same.10 11

In our study there was a significant difference in LVEDD between the control group and the athletes but not between those athletes with and without LVH, after correction for BSA and heart rate (table 2). As in other studies of rowers’ hearts we found mild LVH in 58% of these upper body athletes. When we compared the new stiffness index directly between those athletes with and without LVH, there was no significant difference (p = 0.68) between them. The wide variation in wall thickness in our study group of rowers is similar to those previously described by Pelliccia et al.12

Index of myocardial stiffness

A key determinant of diastolic filling of the left ventricle is chamber stiffness.13 Conventional E/A ratios are considered a measure of diastolic function. Stiffer ventricles show incomplete relaxation, with a greater percentage of filling occurring during the late diastolic phase A. However, this measure is affected by preload and afterload and pseudonormalises as the degree of diastolic dysfunction progresses to a restrictive physiology.14 To overcome these shortcomings, measurement of tissue Doppler diastolic velocities has been developed.1 14 15 They have been shown to be independent of haemodynamic loading conditions and a more robust method of assessing diastolic function. We have previously demonstrated that the stiffness index can differentiate physiological from pathological LVH by the presence of increased LV stiffness and reduced compliance.4 In the study by Spirito et al incorporating 947 elite athletes representing 27 different sports, rowing was ranked first in the likelihood of training-induced increase in LV wall thickness and seventh in the likelihood of training-induced increase in LV cavity size.2 The use of our new stiffness index in this particular study has shown that the key distinguishing feature of intense training is a reduction of myocardial stiffness despite the development of increased wall thickness.

An invasive right heart cardiac catheterisation study measuring ventricular pressure volume curves in 12 elderly endurance athletes compared with 12 age-matched healthy sedentary subjects found that the amount of myocardial stiffness was greater in elderly sedentary subjects than in athletic subjects.16 These investigators concluded that lifelong endurance exercise could prevent the stiffening of heart muscle that is typically caused by ageing. This is in keeping with our finding of a reduced index of stiffness in the healthy endurance rowers compared with controls. Royse et al showed that patients undergoing cardiac surgery demonstrated an increase in end-diastolic stiffness with increasing heart rates.17 Using our new stiffness index, we found that differences in diastolic stiffness between control subjects and rowers remain significant even after correction for heart rate.

The variable physiological response of the heart to training in our small subgroup of those with and without LVH might indicate that other factors (such as genetic) affect the structural heart alterations in elite rowers.

Future applications

Strain rate imaging may provide a future method of assessing myocardial stiffness as it is less influenced by cardiac translation.18 Future improvements in signal-to-noise ratios and refinement of incident beam angle-dependency would potentially make strain rate imaging clinically useful in the measurement of myocardial stiffness which may complement our new stiffness index.

Study limitation

We used M-mode measurements rather than volumetric quantification to assess LVEDD. Although M-mode measurements use geometric assumptions of LV shape and can be limited by the presence of regional wall motion abnormalities, all the subjects in our study were healthy with no evidence of abnormal wall motion.


Elite athletes exhibit reduced myocardial stiffness owing to intense training, irrespective of the development of LVH and dilatation. An understanding of this physiological change may enable more accurate and earlier diagnosis of an abnormal response to high-intensity training.


View Abstract


  • Competing interests: None declared.

  • Ethics approval: Ethics approval was obtained.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles