Parallel increases in systolic and diastolic function characterize the cardiac responses to progressive exercise (29). In a healthy, untrained subject, these myocardial changes serve to maintain a constant stroke volume and ventricular filling volume (after early increases from postural effects), respectively, as systolic and diastolic periods are compromised by increasing HR during upright exercise (25). The extent that enhancement of these functional responses to exercise may contribute to the superior aerobic fitness of trained athletes remains controversial.
Recent investigations have confirmed early studies indicating that the greater maximal stroke volume, cardiac output, and aerobic power of trained athletes compared with those of nonathletes are accounted for by a generalized expansion of the cardiovascular system (augmented ventricular size, blood volume) (11,20,34). These reports imply that variations in magnitude of myocardial functional response (contractile force, relaxation properties) to exercise do not contribute to individual differences in aerobic fitness. Other studies have suggested instead, however, that elevated aerobic fitness in athletes is associated with superior ventricular diastolic function compared with that in nonathletes (12). By this model, qualitative and quantitative cardiac responses to exercise are unique to trained athletes.
The development of ultrasound techniques during exercise has provided certain insights into this issue (26). Diastolic filling of the left ventricle is dictated by the instantaneous pressure gradient created between "upstream" factors (left atrial pressure, atrial contraction) and those "downstream" (ventricular relaxation properties, compliance, and "suction") (40). As estimated by Doppler ultrasound estimation of transmitral flow velocity (E wave), this mean gradient rises approximately fourfold during a bout of progressive exercise in normal, nonathletic individuals (30). Recent studies using tissue Doppler imaging (TDI), which assesses longitudinal myocardial velocity (E′ wave), indicate that the rate of decline of downstream pressure, or ventricular diastolic function, is the primary determinant of this increased transmitral gradient (29), consistent with findings in animals (5). As workload increases, the upstream filling pressure (as estimated by the ratio of E/E′) normally remains stable (14,29).
Enhanced diastolic function in athletes has been suggested by studies indicating their a) higher values of E wave velocity, as well as other markers of diastolic function (isovolumic relaxation time, ratio of early (E) to late (A) filling velocities) at rest [4,16], b) greater resting TDI-E′ (19,21), and c) a continuous rise in stroke volume during progressive exercise (compared with a plateau in nonathletes) (7,12) associated with increasing ventricular size (39), implying a steady increment of ventricular filling volume. However, other studies have failed to indicate differences between athletes and nonathletes in these same echocardiographic markers (29,34,35,37).
Such reports describe diastolic functional characteristics confined to the resting state, and limited insights have been gained from the few studies that have examined echocardiographic markers of diastolic function in athletes during low-intensity exercise or during recovery (8,17). No previous investigation has directly compared such measures in athletes and nonathletes at maximal exercise.
Less controversy surrounds the effects of aerobic fitness on changes in systolic function during exercise. With few exceptions, responses of myocardial contractility (as assessed by left ventricular ejection fraction or shortening fraction) during exercise have generally been found to be similar in athletes and nonathletes (3,6,7,21,34).
The extent that biological immaturity and/or limited training duration might influence myocardial functional responses to exercise is uncertain, as little information is available in young athletes. Nottin et al. (20) described higher E wave values at rest in 12-yr-old male cyclists compared with untrained boys. Both young athletes and nonathletes demonstrate a plateau in stroke volume during progressive exercise, implying a similar augmentation of diastolic function (27). (As noted above, the progressive rise in stroke volume described in some studies of adult athletes has been considered a reflection of superior diastolic function with prolonged training ). No differences between trained and untrained children have been observed in inotropic response to exercise, as indicated by comparable increases in ventricular shortening fraction (21,35).
To further address these issues, this study was designed to examine changes in systolic and diastolic ventricular function by echocardiographic techniques during maximal upright cycle exercise in highly trained midpubertal males compared with nonathletic adolescents. Specifically, this investigation sought to examine the contribution of these myocardial functional responses to the differences in cardiac capacity (maximal cardiac output) and aerobic fitness (peak V˙O2), which separate these two groups.
Twelve highly trained male adolescents (mean age 14.6 ± 0.8 yr) and 10 nonathletic boys (mean age 15.3 ± 0.5 yr) volunteered for exercise testing. Eleven of the athletic subjects were football (soccer) players from an English Premier League club. They reported an average of 7.0 ± 2.2 yr of training, currently practicing 9.9 ± 1.4 months per year, 5.4 ± 1.9 h·wk−1. On average, these boys had been playing in competitive matches for 6.7 ± 0.7 yr. An additional subject who reported cycle training 4 h·wk−1 was included in the athlete group. The untrained group consisted of 10boys who reported little regular physical activity and limited recreational sports participation. All subjects were in good health and were taking no medications that would influence cardiovascular function. By self-assessment, the two groups were matched for level of sexual maturation (for the athletes: average pubic hair Tanner stage was 3, range 2-4; for the nonathletes: average pubic hair Tanner stage was 3, range 2-4).
Informed written parental permission and subjects' assent were obtained for participation. The study was approved by an institutional research ethics committee.
After determination of weight and height, body composition was assessed for descriptive purposes by air displacement plethysmography (Bod Pod; Life Measurement, Inc, Concord, CA). Lean body mass was derived from the difference between body mass and fat content.
All echocardiographic procedures at rest and during exercise were performed by a pediatric cardiologist (T.R.). Measurements of resting left ventricular dimensions and wall thicknesses were obtained by M-mode echocardiography with two-dimensional guidance (Model HD11; Phillips Medical Systems, Eindhoven, the Netherlands) with subjects in the supine position. Dimensions were averaged from triplicate measurements in the parasternal long-axis view, immediately distal to the tips of the open mitral valve leaflets. Left ventricular end-diastolic dimension (LVED) was recorded from the posterior edge of the ventricular septum to the endocardial surface of the posterior wall, and ventricular septal and posterior wall thickness were measured in diastole, all coincident with the Q wave of the ECG. Left ventricular end-systolic dimension (LVES) was measured as the shortest distance during systole between the trailing edge (posterior side) of the ventricular septum and the posterior wall endocardium. Shortening fraction was calculated as (LVED − LVES)/LVED × 100. All measures were adjusted for body size by the square root ofbody surface area (BSA) (13).
Subjects pedaled in the upright position at a cadence of 60 rpm on an electronically braked cycle ergometer (Excalibur Sport 925900; Lode, Groningen, the Netherlands). Initial and incremental loads of 35 W were applied at 3-min intervals to the point of subject exhaustion, defined as the point when the pedal cadence could no longer be maintained.
Gas exchange values were obtained using standard open circuit techniques (Zan 600; nSpire Health Oberthulba, Germany). Minute ventilation was assessed via pneumotachometer. Before testing, the system was calibrated with known oxygen and carbon dioxide concentrations. Peak V˙O2 was defined as the average of the two highest 20-s mean values at the final stage of exercise. This value was expressed relative to both body mass and lean body mass (LBM). A true maximal test was assumed if subjects demonstrated subjective evidence of exhaustion (hyperpnea, sweating, fatigue) plus either a maximal value for RER > 1.00 or HR > 180 bpm.
HR was assessed by ECG. Stroke volume was estimated by conventional Doppler ultrasound techniques (32). Beat-to-beat measures of blood velocity in the ascending aorta were obtained by a 1.9-MHz continuous-wave transducer directed inferiorly from the suprasternal notch with a sweep speed of 100 mm·s−1. Resultant curves were traced off line to obtain the velocity-time integral (VTI). An average of at least three curves with the consistently highest VTI values was averaged and multiplied by the resting aortic valve area to obtain stroke volume. Resting aortic valve area was calculated from the mean of three measurements of the aortic diameter (valve hinge points with leaflets fully open) in the parasternal long-axis view with the subject seated at rest on the cycle ergometer. The aortic valve cross-sectional area was considered to be circular in the sitting position, and changes in mean aperture during the ejection period with increasing work were assumed to be minimal (32). Acceptable reliability and validity of the Doppler echocardiographic technique for estimating stroke volume with maximal exercise has been documented by the first author (T.R.) (28,31,33) and others (see (32) for review).
Cardiac output (Q˙) was calculated as the product of stroke volume and HR. Cardiac output and stroke volume were both indexed to BSA. Arterial venous oxygen difference was computed as V˙O2/Q˙.
The peak blood flow velocity (v) across the mitral valve (E wave) was estimated by pulse wave Doppler echocardiography at the tips of the open mitral valve leaflets in an apical four-chamber view with a 5.0-MHz transducer. The average of the two to three highest values was averaged for each measurement. The maximal instantaneous transmitral pressure gradient (g) was then calculated from the Bernoulli equation g = 4v 2. (Mitral A wave, indicating inflow velocity during late diastole from atrial contraction, was not considered in this analysis because A wave merges with E beyond low exercise intensities.)
Pulse wave TDI for the determination of diastolic E′ was performed at the lateral aspect of the mitral valve annulus with the same transducer in the apical four-chamber view. Effort was made to maintain alignment between the cardiac vertical axis and the transducer beam. The average of at least three consistently highest values was determined for each measurement. TDI recordings were optimized by having one staff member adjust the transducer position while the other operated the echocardiographic controls. Subjects were asked to avoid exaggerated body movement and maintain a generally upright posture but were not otherwise constrained. Measurements were recorded during spontaneous respirations. E′ values were adjusted for heart size by the absolute value of LVED (1,9). Adjusted TDI-E′ was considered a load-independent indicator of the rate of ventricular myocardial relaxation (tau) (36). E/E′ was calculated as an estimate of left ventricular filling pressure (23).
The battery of measurements of HR, VTI, mitral E, and TDI-E′ was obtained at rest, beginning at 1 min 30 s of each 3-min work stage, and during the final minute of exercise. Arterial blood pressure was recorded in the left arm by the auscultatory method at these same times. Diastolic pressure was defined as muffling of sounds. Mean arterial pressure (MAP) was calculated as 1/3 (systolic − diastolic) + diastolic. Systemic vascular resistance was then computed as MAP/Q˙. Stroke work was calculated as the product of stoke volume and mean arterial pressure, an approximation of true stroke work (stroke volume × the integral of systemic arterial pressure during ventricular ejection [18[).
Systolic ejection time was measured from the VTI curve. Systolic function during exercise was assessed by peak aortic velocity (apex of VTI curve) at maximal exercise as well as the systolic ejection rate (stroke volume/systolic ejection time, adjusted for aortic valve area) (22).
SPSS version 16 was used for statistical analyses (SPSS Inc., Chicago, Il). Findings were expressed as mean ± SD. Anthropometric values, resting ventricular dimensions, and hemodynamic variables at rest and at maximal exercise were compared between the two groups by an independent t-test. The significance of changes in variables during exercise was examined by two-way ANOVA (group × time) with repeated measures, with post hoc comparisons by Bonferroni-corrected t-tests. Statistical significance was defined as P≤ 0.05.
No differences were observed in respective values of athletes and nonathletes for body mass (57.1 ± 8.3 and 57.9 ± 10.8 kg), height (168 ± 6 and 171 ± 0 cm), or BSA (1.64 ± 0.13 and 1.67 ± 0.19 m−2). The athletes had significantly lower percent body fat (9.4 ± 5.0) than the untrained subjects (16.0 ± 5.48; P < 0.05). Greater aerobic fitness in the athletes was indicated by a higher peak V˙O2 per kilogram (57.4 ± 4.8 vs 44.4 ± 6.6 mL·kg−1·min−1) and peak V˙O2 per LBM (63.4 ± 5.3 vs 52.4 ± 5.3 mL·kg−1 LBM·min−1) than the controls, respectively. All subjects satisfied criteria for an exhaustive cycling effort, and no significant group differences were observed in RERmax (1.14 ± 0.05 and 1.12 ± 0.05 for athletes and nonathletes, respectively) or maximal HR (188 ± 8 and 195 ± 12 bpm for athletes and nonathletes, respectively).
Resting echocardiographic measurements of left ventricular dimensions, wall thickness, and contractility are outlined in Table 1. When adjusted for body size (BSA0.5), athletes demonstrated significantly greater chamber dimensions, wall thicknesses, and aortic valve area. No group differences were observed in left ventricular shortening fraction as a marker of global myocardial contractility.
Hemodynamic variables at rest and at maximal exercise for the two groups are provided in Table 2. Values were obtained at peak exercise for all subjects except TDI-E′ in two of the athletes, whose curves lacked sufficient consistency and clarity in the final stage (Fig. 1). At rest, the mitral flow profile included early (E wave) and late (Awave) diastolic filling velocities, corresponding to TDI-E′ and TDI-A′ myocardial velocities, respectively. Beginning during low-intensity exercise, mitral E and A waves fused, as did TDI-E′ and TDI-A′. The terms E and E′ are used here to describe these combined velocities during exercise, respectively.
Athletes demonstrated greater maximal stroke index and cardiac index compared with controls but lower values of systemic vascular resistance at peak exercise. There were no significant group differences in calculations of arterial venous oxygen difference or mean arterial pressure at peak exercise. No significant differences between athletes and nonathletes were observed in markers of systolic function (peak aortic velocity, ejection flow rate) or diastolic function (E, adjusted TDI-E′, or E/E′) at rest or at maximal exercise. Mean values of stroke work and stroke index work at peak exercise were significantly greater in the athletes (113 ± 23 vs 89 ± 23 × 10−2 mL·mm Hg−1, and 69 ± 11 vs 53 ± 11 × 10−2 mL·m−2·mm Hg−1, respectively).
The pattern of change in hemodynamic variables with exercise was also similar in the two groups. Stroke volume rose at the initiation of exercise by 18% and 14% after completing the first workload in the athletes and nonathletes, respectively, followed by stable values to exhaustion (Fig. 2). Intragroup one-way ANOVA with repeated measures indicated no significant differences in stroke index between the first workload and exhaustion in both athletes and nonathletes (two-way ANOVA indicated a significant interaction of stroke index between the two groups but with changes of 1-3 mL; this was considered biologically trivial).
The average calculated transmitral peak instantaneous pressure gradient rose by a factor of 4.6 (from 2.1 mm Hg at rest to 9.6 mm Hg at peak exercise) in the athletes and from 2.1 to 8.9 mm Hg in the nonathletes (a 4.3-fold increase; Fig. 3). This increase reflected the effect of downstream factors because TDI-E′ increased by a factor of 2.27 and 2.32 in the two groups, respectively (Fig. 4), whereas mean E/E′ (a marker of upstream filling pressure) values were similar and remained unchanged with exercise (Fig. 5).
Repeated-measures ANOVA revealed main effects for exercise time for all cardiovascular variables except E/E′, which remained stable throughout testing in both groups. No group × time interaction was observed for any factor except stroke index.
Consistent with traditional findings in adults, the young athletes in this study demonstrated greater values for resting left ventricular dimensions as well as maximal cardiac index, stroke index, and peak V˙O2, expressed relative to both body mass and LBM, compared with untrained subjects. No significant difference was observed between groups in peak arterial venous oxygen difference or in mean peak HR. Thus, the superior peak aerobic power and cardiac pumping capacity of the athletes related entirely to their ability to generate peak stroke volume.
Concomitantly, neither qualitative nor quantitative differences in responses of markers of systolic or diastolic function were observed between the two groups as workload increased. Peak transmitral filling pressure gradient rose similarly in the two groups (by a factor of 4.6 and 4.3 in the athletes and nonathletes, respectively). That this augmented gradient reflected a decay in downstream pressure rather than changes in upstream atrial pressure was indicated by stable and equal E/E′ (indicative of ventricular filling pressures) in both groups and similar augmentation of TDI-E′ (reflecting ventricular relaxation properties). Specific to the central question addressed in this study, then these findings fail to indicate that enhancement of changes in myocardial diastolic function contributed to the superior aerobic fitness of these adolescent athletes.
The pattern of stroke volume as workload rose was similar in the athletes and nonathletes, with a 14%-19% rise at the onset of exercise and then stable values (a "plateau") consequently to the point of subject fatigue. The nature of this pattern provides important information regarding changes in ventricular diastolic function during a bout of progressive exercise. Constant values of stroke volume with increasing work indicate that functional responses are augmented to maintain a stable ventricular filling volume as diastolic period shortens with increasing HR. If diastolic function is further enhanced, ventricular filling volume will increase, and stroke volume will progressively rise as workload increases via a Frank-Starling mechanism. The plateau of stroke volume observed in this study in both groups, then, is consistent with the lack of echocardiographic evidence for enhancement of diastolic function in the trained athletes compared with nonathletes.
Some reports have indicated, however, that stroke volume does not plateau during a progressive exercise test in highly trained adult endurance athletes but instead rises progressively (see  for review). Although such findings suggest superior diastolic function in these athletes, an equal number of studies has demonstrated, to the contrary, a stroke volume plateau in highly fit endurance athletes. Methodological differences may explain these variations in reports of stroke volume patterns in athletes (27).
Previous studies comparing echocardiographic markers of diastolic function at rest between adult athletes and nonathletes have provided conflicting information. Mitral E and E/A, markers of transmitral pressure gradient, have been reported to be similar or greater in athletes (4,16). In the latter reports, the issue may be clouded by failure to account for the resting bradycardia typically evident among the athletes because these echocardiographic variables are influenced by HR (10,15,24,38). For instance, Gilliam et al. (10) found that an increase in paced HR from 60 to 85 bpm resulted in a decrease in average E from 0.48 to 0.30 m·s−1, increased A from 0.68 to 0.79 m·s−1, and decline in E/A ratio from 0.70 to 0.30. (It is possible, however, that changes with artificial pacing may not reflect those occurring with increased HR with exercise.)
TDI-E′ has been reported greater in adult athletes than in nonathletes at rest (19,21), but others have not confirmed this difference (37). In this case, the pitfall is failure to correct values for the greater absolute ventricular size of the athletes because E′ is directly and linearly related to ventricular linear dimension (1) and LVED (9).
Data addressing the more specific question of diastolic function in athletes during exercise are meager. The present study is the first to compare mitral E, TDI-E′, and E/E′ during peak upright exercise in athletes and nonathletes. Previous studies have reported findings during low-grade supine exercise or postexercise (17). DiBello et al. (8) described changes in mitral E velocity during semisupine maximal exercise in adult distance runners and untrained subjects. Average values were similar at rest (0.82 and 0.80 cm·s−1 for athletes and controls, respectively) and also at all submaximal work levels. At peak exercise, the average E velocity was greater in the athletes (1.54 vs 1.42 cm·s−1), but because their endurance time was also greater (5.05 ± 0.9 vs 4.28 ± 0.7 min), it was not possible to conclude whether the higher mitral gradient was a cause or a result of superior fitness.
By two markers of systolic function, maximal peak aortic blood velocity and ejection rate (adjusted for aortic valve area), this study confirmed previous reports indicating no differences in myocardial contractility in the athletes and nonathletes during progressive exercise. Improvement in "contractility" is here defined in a global sense, indicating greater speed, acceleration, and force of myocardial fiber shortening. Such accentuated function may reflect a decline in peripheral vascular resistance, increased HR, and/or sympathetic and catecholamine stimulation, factors which all become accentuated during a progressive exercise test. Although it has not been possible to decipher the relative contributions of each to overall augmentation of ventricular systolic functional responses to exercise, findings in this study and in previous studies generally indicate that such changes are no different in athletes and nonathletes (6).
This study indicated that given ultrasound-experienced personnel and close attention to technical detail, assessment of mitral flow velocity and TDI is feasible during upright cycle exercise, at least in young, nonobese subjects. Findings reported by Bougault et al. (2) indicate acceptable reproducibility of this technique during semisupine exercise in 19- to 34-yr-old nonobese men. Test-retest correlation for TDI-E′ velocities at peak exercise recorded at the lateral mitral annulus was r = 0.87, with a coefficient of variation of 4.8%. The mean difference in E′ in the two tests was 0.17 cm·s−1, with 95% limits of agreement of −6.12 and ±6.46 cm·s−1. Test-retest coefficient of variation for mitral E was 2.8% at peak exercise.
Establishment of the validity of ultrasound techniques in assessing ventricular function during maximal exercise is challenging, given the lack of "gold standard" method. However, some evidence exists for concurrent validity. In this study, average peak mitral E velocity rose from 72 cm·s−1 at rest to 152 cm·s−1 at maximal exercise, a 2.1-fold increase. In a previous study involving 7- to 12-yr-old boys (29), values increased from 76 to 143 cm·s−1 at maximal exercise (1.9-fold rise). In the study by Bougault et al. (2), mitral E rose from 76 cm·s−1 at rest to 159 cm·s−1 at maximal exercise (a 2.1-fold increase).
The increase in TDI-E′ in this study from 13 to 30 cm·s−1 (2.3-fold rise) at peak exercise is also consistent with previous reports. Bougault et al. (2) described a rise from 19 to 33 cm·s−1 (1.7-fold increase) at exhaustion during semisupine exercise, whereas Rowland et al. (29) reported an increase from 6.3 to 12.1 cm·s−1 in mean (instead of peak) TDI-E′ values with upright cycling (1.9-fold increase).
The findings in this study contribute to the growing body of research addressing the basic mechanisms that underlie the superior cardiac and aerobic fitness in trained athletes. Specifically, the findings in this investigation of midpubertal males with an average of 7 yr of competitive training and superior peak V˙O2 are consistent with a conceptual model by which maximal stroke volume-the essential factor defining maximal aerobic power between athletes and nonathletes-is an expression of ventricular volume at rest and during exercise. By this concept, higher Q˙ max and peak V˙O2 reflect factors that augment ventricular size in trained individuals (plasma volume, resting bradycardia) rather than inotropic and lusitropic functional characteristics of the myocardium unique to the trained athlete.
It is important to recognize that conclusions from this study are limited to a group of highly fit but not elite (i.e., internationally competitive) adolescent athletes. Whether qualitative differences in ventricular functional responses to exercise might be observed in more highly trained adults with greater duration of training or whether such responses might be influenced by factors such as gender, type of sport, or aging, deserves attention in future research investigations.
Using echocardiographic techniques, this is the first study to compare directly ventricular systolic and diastolic functional responses to maximal exercise in trained athletes and nonathletes. It will be important in future investigations to use such techniques to assess cardiac functional characteristics at a high-intensity exercise, thereby avoiding the confounding effects of HR and ventricular size, which may influence markers of diastolic and systolic function in the resting state.
In doing so, certain potential limitations need to be addressed. 1) Further research is needed regarding the validity and reliability of TDI and mitral flow velocity during exercise. Specifically, the relationship of TDI markers to ventricular function during exercise needs to be confirmed. 2) Optimal techniques for measuring TDI variables during exercise should be established. 3) The generalization of findings in this study to subjects with greater duration and intensity of endurance training needs to be examined.
The authors are indebted to Philips Healthcare Ultrasound Division, Surrey, United Kingdom, and Gillian Nash, cardiac ultrasound sales specialist, for their kind assistance in providing the echocardiographic equipment for this study.
There are no personal or financial relationships with other people or organizations that could represent potential conflicts of interest relating to the authors or their respective institutions.
This study was conducted with no external funding.
The results of this study do not constitute endorsement by ACSM.
1. Batterham A, Shave R, Oxborough D, Whyte G, George K. Longitudinal plane colour tissue-Doppler myocardial velocities and their association with left ventricular length, volume, and mass in humans. Eur J Echocardiogr
2. Bougault V, Nottin S, Doucende G, Obert P. Tissue Doppler imaging reproducibility during exercise. Int J Sports Med
3. Brandao MUP, Wajngarten M, Rondon E, Girogi MCP, Hironaka F, Negrfao CE. Left ventricular function during dynamic exercise in untrained and moderately trained subjects. J Appl Physiol
4. Casa P, D'Andrea A, Galderisi M, et al. Pulsed Doppler tissue imaging in endurance athletes: relation between left ventricular preload and myocardial regional diastolic function. Am J Cardiol
5. Cheng C-P, Igarashi Y, Little WC. Mechanism of augmented rate of left ventricular filling during exercise. Circ Res
6. Colan SD, Sanders SP, Borow KM. Physiologic hypertrophy: effects on left ventricular systolic mechanics in athletes. J Am Coll Cardiol
7. Crawford MH, Petru MA, Rabinowitz C. Effect of isotonic training on left ventricular volume during upright exercise. Circulation
8. Di Bello V, Santoro G, Talarico L, et al. Left ventricular function during exercise in athletes and in sedentary men. Med. Sci. Sports Exerc
9. Eidem BW, McMahon CJ, Cohen RR, et al. Impact of cardiac growth on Doppler tissue imaging velocities: a study in healthy children. J Am Soc Echocardiogr
10. Gilliam LD, Homma S, Novick SS, Rediker DE, Eagle KA. The influence of heart rate on Doppler mitral inflow patterns [abstract]. Circulation
. 1987;76(Suppl IV):IV-123.
11. Gintzon LE, Conant R, Brizendine M, Laks MM. Effect of high intensity aerobic training on left ventricular volume during maximal upright exercise. J Am Coll Cardiol
12. Gledhill N, Cox D, Jamnik R. Endurance athletes' stroke volume does not plateau: major advantage is diastolic function. Med Sci Sports Exerc
13. Gutgesell N, Rembold CM. Growth of the human heart relative to body surface area. Am J Cardiol
14. Ha JW, Oh JK, Pellikka PA, et al. Diastolic stress echocardiography: a novel noninvasive diagnostic test for diastolic dysfunction using supine cycle exercise Doppler echocardiography. J Am Soc Echocardiogr
15. Harrison MR, Clifton GD, Pennell AT, DeMaria AN. Effect of heart rate on left ventricular diastolic transmitral flow velocity patterns assessed by Doppler echocardiography in normal subjects. Am J Cardiol
16. Hoogsteen J, Hoogeveen A, Schaffers H, Wijn PF, van der Wall EE. Left atrial and ventricular dimensions in highly trained cyclists. Int J Cardiovasc Imag
17. Huonker M, Konig S, Keul J. Assessment of left ventricular dimensions and functions in athletes and sedentary subjects at rest and during exercise using echocardiographic, Doppler sonography and radionuclide ventriculography. Int J Sports Med
. 1996;17(3 suppl):S173-79.
18. Katz AM. Physiology of the Heart
. 3rd ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2001. p. 141-148.
19. Koc M, Bozkurt A, Akpiner O, Ergen N, Acarturk E. Right and left ventricular adaptation to training determined by conventional echocardiography and tissue Doppler imaging in young endurance athletes. Acta Cardiol
20. Nottin S, Nguyen L-D, Terbah M, Obert P. Left ventricular function in endurance-trained children by tissue Doppler imaging. Med Sci Sports Exerc
21. Nottin S, Vinet A, Stecken F, Obert P. Central and peripheral cardiovascular adaptations to exercise in endurance trained children. Acta Physiol Scand
22. Nutter DO, Noble RJ, Hurst VW. Peak aortic flow and acceleration as indices of ventricular performance in the dog. J Clin Lab Med
23. Ommen SR, Nishimura RA, Appleton CP. Clinical utility of Doppler echocardiography and tissue Doppler imaging in the estimation of left ventricular filling pressures. Circulation
24. Oniki T, Hashimoto Y, Shimizu S, Kakuta T, Yajima M, Numano F. Effect of increasing heart rate on Doppler indices of left ventricular performance in healthy men. Br Heart J
25. Rowland T. Circulatory responses to exercise. Are we misreading Fick? Chest
26. Rowland T. Circulatory responses to progressive endurance exercise: insights from echocardiography. Sports Med
27. Rowland T. Stroke volume response to progressive exercise in endurance athletes: a critical assessment. Sports Med
. in press.
28. Rowland T, Blum JW. Cardiac dynamics during upright cycle exercise in boys. Am J Hum Biol
29. Rowland T, Heffernan K, Jae SY, Echols G, Fernhall B. Tissue Doppler assessment of ventricular function during cycling in 7- to 12-yr-old boys. Med Sci Sports Exerc
30. Rowland T, Mannie E, Gawle L. Dynamics of left ventricular filling with exercise. Chest
31. Rowland T, Melanson E, Popowski B. Test-retest reproducibility of maximal cardiac output by Doppler echocardiography. Am J Cardiol
32. Rowland T, Obert P. Doppler echocardiography for estimation of cardiac output with exercise. Sports Med
33. Rowland T, Popowski B. Comparison of bioimpedance and Doppler cardiac output during exercise in children [abstract]. Pediatr Exerc Sci
34. Rowland T, Roti M. Cardiac responses to progressive upright exercise in adult male cyclists. J Sports Med Phys Fitness
35. Rowland T, Unnithan V, Fernhall B, Baynard T, Lange C. Left ventricular response to dynamic exercise in young cyclists. Med Sci Sports Exerc
36. Sohn D-W, Chaiw I-H, Lee D-J. Assessment of mitral annulus velocity by tissue Doppler imaging in the evaluation of leftventricular diastolic function. J Am Coll Cardiol
37. Schmidt-Truckass A, Schmid A, Haussler D, Huber G, Huonker M, Keul J. Left ventricular wall motion during diastolic filling inendurance-trained athletes. Med Sci Sports Exerc
38. Smith SA, Stozner JE, Russell AE, Sheppard JM, Aylward PE. Transmitral velocities measured by pulsed Doppler in healthy volunteers: effects of acute changes in blood pressure and heart rate. Br Heart J
39. Warburton DER, Haykowsky MJ, Quinney HA, Blackmore D, Teo KK, Human DP. Myocardial response to incremental exercisein endurance trained athletes: influence of heart rate, contractility, and the Frank-Starling effect. Exp Physiol
40. Yellin EL, Nikolic S, Frater RWM. Left ventricular filling dynamics and diastolic function. Prog Cardiovasc Dis
Keywords:©2009The American College of Sports Medicine
AEROBIC FITNESS; HEART FUNCTION; FITNESS TESTING; ULTRASOUND