It is well established that chronic endurance exercise training induces morphological left ventricular (LV) adaptation, a phenomenon commonly called athlete’s heart (2,20,28). Exercise-induced right ventricular (RV) adaptation has received less attention, largely because of the irregular shape of the RV and the associated difficulty in assessing structure and function using traditional echocardiographic imaging methods (15,18,21). Cross-sectional echocardiographic studies evaluating RV adaptation to exercise suggest that eccentric RV hypertrophy is evident in athletes compared with controls (4,6,7,16,30). This is supported by two magnetic resonance imaging (MRI) studies that also report that the eccentric morphology in the right ventricle is balanced with the left ventricle (26,27). The effect of resistance training on the right ventricle is less well known with limited evidence using echocardiography (2). An MRI study of “anaerobic” (track sprinters) versus “aerobic” athletes (marathon runners) and controls reported pronounced RV enlargement in the anaerobic athletes, which runs contrary to most echocardiographic athlete–control studies (3).
The disparity concerning RV adaptation to exercise may be due to several limitations. First, most previous studies have used a cross-sectional design, which restricts inference about causality and is limited by selection bias, poor detail of training exposure, and inconsistent approaches to scaling or indexing of cardiac data (20). Longitudinal training approaches with repeated measures within individuals alleviate these issues, but few well-controlled studies have been conducted regarding the right ventricle, and these have used echocardiography (2,19). Second, measures of RV morphology using echocardiography are largely one- or two-dimensional (e.g., diameter at the RV outflow tract). Using these constructs as substitutes of three-dimensional RV structure oversimplifies the complexity of the RV shape and has a high degree of variability (8). Cardiac MRI, however, is considered the gold standard for RV structural assessment, with a reproducibility of 4.2% and 7.8% for the RV end-diastolic volume (RVEDV) and RV mass, respectively (11,18). To date, no studies have used MRI to evaluate the effect of different training interventions on RV morphological adaptations.
We undertook a prospective, longitudinal, randomized study comparing the effect of endurance and resistance exercise training on RV morphological and functional adaptation in young, healthy humans, using cardiac MRI. We recently published data suggesting a lack of concentric LV hypertrophy after resistance training (28) and therefore hypothesized that endurance training would induce RV eccentric hypertrophy, symmetrical with the left ventricle, whereas resistance training would not.
Twenty-seven healthy, untrained male subjects were recruited after a thorough prescreening, including detailed medical history, physical examination, standard blood panels, and physical activity questionnaire. None of the subjects experienced diabetes, cardiovascular disease, used any medication, smoked, or participated in recreational activity >3 h·wk−1. Subjects were randomly assigned to either an endurance (E) or resistance (R) training group using a randomization scheme generated using online software (www.randomization.com). Before any data collection, three subjects withdrew from the E group citing work and/or study commitments. Another subject from the R group voluntarily withdrew because of interstate work transfer (n = 1). We adopted a per protocol analysis of subjects who complied with the training, such that the reported data represent the remaining 23 subjects (E = 10, R = 13). This study complies with the Declaration of Helsinki, and the Human Research and Ethics Committee of the University of Western Australia approved the experimental protocol. All subjects provided written, informed consent before participating in the study.
Subjects participated in a 24-wk R or E training program in accordance with a parallel group randomized controlled trial. Baseline measures were taken before program commencement and included body composition assessment using dual-energy x-ray absorptiometry, aerobic fitness using a graded exercise test, muscular strength using the one-repetition maximum (1RM) protocol, RV morphological assessment with MRI, and RV functional measurement using myocardial speckle tracking echocardiography. Repeat measures were taken after 24 wk of exercise training. In summary, the E intervention consisted of a progressively overloaded program of walking/jogging/running, divided into three training phases for the 24-wk period. The focus of the periodized R program was Olympic weightlifting with incorporated assistance exercises (e.g., dead lift, squat, bench press, overhead press) to develop overall strength and technique. Subjects attended three 1-hr sessions per week with individualized relative intensities for the interventions so that subjects were exercising at prescribed percentages of peak oxygen consumption (V˙O2peak) and 1RM, respectively, ranging from 40% to 100% depending on the training phase. A more detailed description of the exercise-training programs used is published elsewhere (28).
Before the graded exercise test, subjects underwent a whole-body dual-energy x-ray absorptiometry assessment (Lunar Prodigy; GE Medical Systems, Madison, WI) to determine body composition, specifically total fat mass, total lean body mass and body fat percentage.
Graded exercise tests were performed on a treadmill to determine V˙O2peak as reported previously (28). Peak oxygen consumption was expressed as the summation of the four highest consecutive 15-s V˙O2 values in each workload.
Maximal upper and lower body strength were determined as described previously (28), and combined muscular strength was expressed as the summation of 1RM scores for the bench press and squat.
Cardiac morphology: MRI
RV morphology was assessed using a 1.5-T cardiac MRI unit (Magnetom Espree; Siemens, Erlangen, Germany). Using TrueFISP, images were acquired with the subject in a supine position with a posterior phased array spine coil and an anterior flexible phased array body surface coil. ECG-gated multiplane breath-hold sequences were applied with standard cardiac imaging planes obtained. For all sequences, the breath-hold times were between 5 and 20 s, dependent on the subject’s heart rate. Images of the right ventricle were collected in the short axis plane, perpendicular to the ventricular septum with 10–12 slices acquired (repetition time = 37.68, echo time = 1.29, flip angle = 70°–80°, field of view = 320–350 mm, slice thickness = 6 mm, interslice gap = 4 mm, resolution = 256 × 166, bandwidth = 930). Images of the left ventricle were also acquired, details of which have been described previously (28).
Cardiac MRI analysis was performed using specialized commercially available software (ARGUS; Siemens) by an observer blinded to the subject group and time point. Analyses were independently repeated and confirmed by an experienced radiologist (CM), also blinded to the subject group and time point. Laboratory-specific reliability data for repeated cardiac MRI measures (coefficient of variation = 1.1%–2.5%, intraclass correlation = 0.95–0.97) are in agreement with previous literature (1). To overcome difficulties in determining the right ventricle in the basal two slices, instructions for tracing the endo- and epicardial borders were taken from the methodology described by Prakken et al. (24). Short axis cine loops were inspected for end systole, which was defined as the frame with the smallest ventricular cavity. In the most basal end-diastolic slice, an RV contour was drawn only if it was visible for a minimum of three phases in diastole, whereas the visible RV contour was always traced in the most basal end-systolic slice. RV mass was determined by subtracting the volumes between the epi- and the endocardial borders, multiplying by slice thickness and specific density of the myocardium and summing the slices of this area during end diastole and end systole, with the mass at end diastole being reported. As the epicardial border overlapped both the septal part of the endocardial border and at the valve planes, the RV mass was calculated from the RV lateral wall only (Fig. 1). The RVEDV, the RV end-systolic volume, and the derived RV ejection fraction, stroke volume, and cardiac output were calculated.
Myocardial function: speckle tracking echocardiography
We performed myocardial speckle tracking of the right ventricle to provide a corroborative technique that has previously been used in exercise studies and disease for the assessment of RV function (21). These data provide complementary information to RV morphology, specifically on longitudinal deformation, which is not readily available using MRI. As it has been demonstrated, in the left ventricle, that longitudinal changes occur before global indices, it could be argued that myocardial speckle tracking would provide an early marker of RV improvement. Subjects lay in the left lateral decubitus position and a single experienced sonographer obtained images using a commercially available ultrasound system (Vivid I; GE Medical, Horton, Norway). Image acquisition followed a standard echocardiographic protocol. From an apical four-chamber orientation, a focused RV view was achieved with lateral movement of the transducer (25). Images were optimized with gain, dynamic range, and depth to ensure optimal endocardial delineation. The focal point was positioned midcavity to reduce the effect of beam divergence, and frame rate was maintained as close to 90 fps as possible ensuring consistency from all acquisitions. Cine loops for three cardiac cycles were recorded in a DVD in a raw DICOM format, and all data were analyzed offline by a single observer, who had no knowledge of group allocation or time point, using specialized software with specific speckle tracking capabilities (EchoPAC; GE Healthcare, Norway) in accordance with Oxborough et al. (22) to determine RV peak longitudinal strain and strain rate (SR). A region of interest was placed around the RV lateral wall from base to apex for the myocardial speckle tracking analysis. Peak RV longitudinal strain and SR were assessed in three wall segments: basal, midwall, and apical (Fig. 2). This was averaged to provide global measurements of peak strain and peak SR recorded in systole, early diastole, and late diastole with data averaged for a minimum of three continuous cardiac cycles.
The effectiveness of each training intervention for measures of RV cardiac morphology and function, aerobic fitness, body composition, and muscular strength was analyzed using an ANCOVA model (31). This is the most appropriate analysis for a small, randomized trial as this adjusts for chance imbalance at study entry in any outcome variable. The delta or change score of all outcome measures was used as the dependent variable, group as the independent variable, and prestudy data as the covariate. Data are reported as mean ± SD as well as 95% confidence intervals (95% CI). All statistical analyses were performed using PASW Statistics for Windows version 18 (SPSS Inc., Chicago, IL) and Excel (Microsoft Office Excel 2007).
Subject characteristics and training
There were no differences between changes in fat mass (kg) in either group (Table 1), but lean mass increased to a similar extent in both R (+2.1 kg, 95% CI = 1.5–3.2) and E training (+1.3 kg, 95% CI = 0.3–2.3). ANCOVA revealed a significant difference in training response between groups for V˙O2peak, which was increased after E (+3.7 mL·kg−1·min−1, 95% CI = 0.9–6.5) but not R training (−0.2 mL·kg−1·min−1, 95% CI = −2.6 to 2.3; Table 1). Although strength was increased in both groups, ANCOVA revealed a significantly greater change in combined strength after R training (Table 1).
RV morphology and function
All RV morphological data are presented in Table 2. RV mass increased 2.7 g (95% CI = −0.4 to 5.8 g) after E training and 1.4 g (95% CI = −1.3 to 4.1 g) after R training. This change score was not different between groups (P > 0.05). A minor increase in RV mass after training persisted even after scaling for body surface area, but again there was no difference in the change score between groups. The change in RVEDV was 13.8 mL (95% CI = 1.9–25.7 mL) after E and 3.9 mL (95% CI = −6.5 to 14.3 mL) after R training. Although larger after E training this was not significantly different between groups. The change in RV ejection fraction, stroke volume, and cardiac output after training was negligible in both groups. Changes in RV longitudinal peak RV strain and SR were small and not different between groups (Table 3). LV results in these subjects are published elsewhere (28). The LV-to-RV ratios for mass, EDV, stroke volume, and ejection fraction were unchanged following training (Fig. 3).
We demonstrate for the first time using MRI that RV mass was increased after 6 months of progressive and intense E or R training in young, healthy, and previously untrained male subjects. The change in RV mass in absolute terms was small and not significantly different between groups, although the relative response was greater after E (9%) than R training (3%). A similar pattern emerged in relation to changes in RVEDV after training. The ratios of LV-to-RV morphology and function were unchanged after training, indicative of a modest but balanced adaptation to the exercise stimulus. These cardiac structural adaptations occurred in the absence of any significant change in global myocardial function.
These data are in partial agreement with the hypothesis that exposure to chronic endurance exercise training induces an eccentric-type hypertrophy of the right ventricle. The current data provide support for previous cross-sectional echocardiographic studies in athletes that reported increased end-diastolic diameter of the RV outflow tract (6,13,16,21,29,30) and MRI studies that have reported increased RV mass and RVEDV (26,27) in athletes compared with controls subjects. Similarly, a longitudinal study of endurance-trained athletes demonstrated further increases in RV parameters after 3 months of unsupervised training (2). It is possible that the hemodynamic overload due to the repetitive episodic increases in preload, which occur during endurance activity (20), may play a role in the mechanistic cascade for RV adaptation to E exercise.
Scharhag et al. (27) reported increased LV and RV mass of 36% and 37%, respectively, in elite endurance athletes compared with matched controls using MRI. Likewise previous echocardiography studies have described greater differences in RV morphology between athletes and controls than those seen after training in the current study. Although the between group difference in LV and RV mass demonstrated by Scharhag et al. (27) exceeds the 9% increase in RV mass we observed with E exercise, a longer history of training in the athletic cohort could largely account for this difference. Longer training studies to approach athletic training loads would be valuable but are logistically difficult.
The current data support a balanced response of the left and right ventricles to training that has been observed in some cross-sectional athlete–control comparisons (26,27). In contrast to our findings, some descriptive studies have reported asymmetrical hypertrophy of the right ventricle of athletes (18,23). A recent multimodal imaging study found disproportionate RV structural adaptation in middle-age endurance athletes (mean age = ∼36 yr) versus controls (17). The authors suggest this asymmetry was due to a disproportionate end-systolic wall stress imposed on the right ventricle compared with the left ventricle during an acute bout of exercise. We did not assess wall stress in response to acute exercise (either E or R training) in the present study, but this may be worth assessing in future training studies. Arguably, the assessment of end-diastolic wall stress (an indicator of preload) may be a more viable method of assessing the stimulus for adaptation to ventricular volume overload. In addition, the mean age of the subjects in our study is ∼10 yr younger than those of La Gerche et al. (17), which may partially account for the differences we observed.
Cross-sectional athlete–control studies (either MRI or echocardiography) are equivocal in their description of a concentric hypertrophy in resistance-trained subjects (5,9,23,32). The current study supports the contention that R training has a somewhat smaller effect on RV morphology when compared with E training. To our knowledge, Baggish et al. (2) conducted the only other longitudinal evaluation of RV structure comparing endurance and resistance exercise training. Although this 3-month echocardiographic study was observational in nature, with no supervised, structured, or controlled interventions, “endurance” collegiate rowers significantly enhanced RV dimension by 6% from baseline, whereas “strength” collegiate American footballers had no alteration in RV diameter. These findings are largely supported by the current MRI data. The smaller RV adaptation we and others observed in response to R training may be explained by the limited magnitude and time duration of exposure to increased RV loading within R training that is characterized by intermittent activity that does not place a constant, steady-state hemodynamic overload on the heart (20). Interestingly, in the hemodynamic stimulus thought to promote cardiac adaptation to R training (10), an increased LV end-systolic wall stress may not even occur during R exercise (12), which raises the question whether the hemodynamic stimulus is sufficient to induce cardiac morphological adaptation.
Using traditional global functional indices (RV ejection fraction and stroke volume) or the novel application of myocardial speckle tracking echocardiography, we have demonstrated no significant change in RV myocardial function in a young, healthy men undergoing 6 months of intense and progressive E or R training. Further, the lack of functional adaptation occurred despite modest changes in RV morphology after exercise training. As global strain derived from myocardial speckle tracking is free from angle dependence, some of the geometric limitations of ultrasound are overcome (22), and it can be successfully applied to multimodal assessment of the right ventricle.
Magnetic resonance imaging is considered the gold standard for structural assessment of the right ventricle (18) yet is subject to some limitations. First, it is difficult to distinguish the RV inflow area with the tricuspid valve from the outflow tract and pulmonary valve in the most basal RV slice (11,14). Although long-axis and biplane methods may be an alternative for assessing RV volume (1), these methods rely heavily on geometric assumption of ventricular shape. By adopting a standardized method of tracing the epi- and endocardial borders, we have attempted to minimize the error associated with movement of the atrial–ventricular plane during diastole. Furthermore, delineation of the endocardial border can be difficult because of the thin RV lateral wall and increased trabeculations, particularly during systole (11). The combination of a descriptive analysis protocol (24) and the steady-state precision technique (TrueFISP) used in our imaging procedure allowed for improved contrast between myocardium and blood pool to reduce this limitation. We recruited only male subjects for this study. Previous research suggests that the cardiac morphology of female athletes may adapt differently compared with their male athletic counterparts (2). Lastly, we acknowledge that exposure to a longer training stimulus may evoke greater adaptive responses and thus cannot rule out some adaptation because of R exercise under such circumstances. Our findings should be judiciously compared with previous cross-sectional studies of lifelong endurance or resistance athletes. However, we maximized the training responses observed by recruiting previously untrained subjects and closely supervising the intensive training sessions. In summary, additional longitudinal approaches, particularly in women, will be required to determine the long-term health implications and gender differences of exercise training on RV adaptation and to ascertain any dose-response relationship.
In conclusion, a 24-wk intensive, supervised, and controlled E exercise-training program resulted in a modest but increase in RV mass and RVEDV, suggestive of eccentric remodeling. This was similar to the pattern of change observed in the left ventricle. Changes in RV morphology were small after R training but not significantly different from those observed with E training. No significant changes in global or regional RV function at rest were observed because of either E or R training.
The authors thank Ms. Kate Taylor for her assistance with echocardiography support and Mr. Philip Watson and Ms. Kylie Williams for their MRI technical support.
DJG was supported by funding from the National Heart Foundation of Australia and the Australian Research Council. There are no disclosures or conflict of interest declared for each author. The findings of the present study do not constitute endorsement by the American College of Sports Medicine.
ALS and HHC performed data collection, exercise training, data analysis and interpretation, manuscript drafting and revision, and approval of the final manuscript. LHN, CPM, and DO were responsible for data collection, data analysis and interpretation, manuscript drafting and revision, and approval of the final manuscript. KPG and DJG conceived and designed the study, interpretation and analysis of data, manuscript drafting, and revision and final approval of the manuscript.
1. Alfakih K, Plein S, Thiele H, Jones T, Ridgway JP, Sivananthan MU. Normal human left and right ventricular dimensions for MRI as assessed by turbo gradient echo and steady-state free precession imaging sequences. J Magn Reson Imaging
. 2003; 17 (3): 323–329.
2. Baggish AL, Wang F, Weiner RB, et al.. Training-specific changes in cardiac structure and function: a prospective and longitudinal assessment of competitive athletes. J Appl Physiol
. 2008; 104 (4): 1121–1128.
3. D’Andrea A, Caso P, Sarubbi B, et al.. Right ventricular myocardial adaptation to different training protocols in top-level athletes. Echocardiography
. 2003; 20 (4): 329–336.
4. D’Andrea A, Caso P, Severino S, et al.. Different involvement of right ventricular myocardial function in either physiologic or pathologic left ventricular hypertrophy: a Doppler tissue study. J Am Soc Echocardiogr
. 2003; 16 (2): 154–161.
5. D’Andrea A, Limongelli G, Caso P, et al.. Association between left ventricular structure and cardiac performance during effort in two morphological forms of athlete’s heart
. Int J Cardiol
. 2002; 86 (2–3): 177–184.
6. D’Andrea A, Naeije R, D’Alto M, et al.. Range in pulmonary artery systolic pressure among highly trained athletes. Chest
. 2011; 139 (4): 788–794.
7. D’Andrea A, Riegler L, Golia E, et al.. Range of right heart measurements in top-level athletes: The training impact. Int J Cardiol
. 2011; 1–10; [epub ahead of print July 5, 2011].
8. Foale R, Nihoyannopoulos P, McKenna W, Klienebenne A, Nadazdin A, Rowland E. Echocardiographic measurement of the normal adult right ventricle. Brit Heart J
. 1986; 56: 33–44.
9. George KP, Batterham A, Jones B. Echocardiographic evidence of concentric left ventricular enlargement in female weight lifters. Eur J Appl Physiol
. 1998; 79: 88–92.
10. George KP, Wolfe LA, Burggraf GW. The “athletic heart syndrome.” A critical review. Sports Med
. 1991; 11 (5): 300–330.
11. Grothues F, Moon JC, Bellenger NG, Smith GS, Klein HU, Pennell DJ. Interstudy reproducibility of right ventricular volumes, function, and mass with cardiovascular magnetic resonance. Am Heart J
. 2004; 147 (2): 218–223.
12. Haykowsky MJ, Taylor D, Teo K, Quinney A, Humen D. Left ventricular wall stress during leg-press exercise performed with a brief Valsalva maneuver. Chest
. 2001; 119 (1): 150–154.
13. Henriksen E, Kangro T, Jonason T, et al.. An echocardiographic study of right ventricular adaptation to physical exercise in elite male orienteers. Clin Physiol
. 1998; 18 (6): 498–503.
14. Hudsmith L, Petersen S, Francis J, Robson M, Neubauer S. Normal human left and right ventricular and left atrial dimensions using steady state free precession magnetic resonance imaging
. J Cardiovasc Magn Reson
. 2005; 7 (5): 775–782.
15. Jurcut R, Giusca S, La Gerche A, Vasile S, Ginghina C, Voigt JU. The echocardiographic assessment of the right ventricle: what to do in 2010? Eur J Echocardiogr
. 2010; 11 (2): 81–96.
16. Kasikcioglu E, Oflaz H, Akhan H, Kayserilioglu A. Right ventricular myocardial performance index and exercise capacity in athletes. Heart Vessels
. 2005; 20 (4): 147–152.
17. La Gerche A, Heidbuchel H, Burns A, et al.. Disproportionate exercise load and remodeling of the athlete’s right ventricle. Med Sci Sports Exerc
. 2011; 43 (6): 974–981.
18. La Gerche A, Taylor AJ, Prior DL. Athlete’s heart
: the potential for multimodality imaging to address the critical remaining questions. J Am Coll Cardiol Img
. 2009; 2 (3): 350–363.
19. Molmen HE, Wisloff U, Aamot IL, Stoylen A, Ingul CB. Aerobic interval training compensates age related decline in cardiac function. Scand Cardiovasc J
. 2012; 46 (3): 163–171.
20. Naylor L, George KP, O’Driscoll G, Green DJ. The athletes heart: a contemporary appraisal of the Morganroth hypothesis. Sports Med
. 2008; 38 (1): 69–90.
21. Oxborough D, Sharma S, Shave R, et al.. The right ventricle of the endurance athlete: the relationship between morphology and deformation. J Am Soc Echocardiogr
. 2012; 25 (3): 263–271.
22. Oxborough D, Shave R, Warburton DER, et al.. Dilatation and dysfunction of the right ventricle immediately after ultraendurance exercise: exploratory insights from conventional two-dimensional and speckle tracking echocardiography. Circ Cardiovasc Imaging
. 2011; 4 (3): 253–263.
23. Perseghin G, Decobelli F, Esposito A, et al.. Effect of the sporting discipline on the right and left ventricular morphology and function of elite male track runners: a magnetic resonance imaging
and phosphorus 31 spectroscopy study. Am Heart J
. 2007; 154 (5): 937–942.
24. Prakken N, Velthuis B, Vonken E, Mali W. Cardiac MRI: standardized right and left ventricular quantification by briefly coaching inexperienced personnel. Open Magn Reson J
. 2008; 1: 104–111.
25. Rudski L, Lai W, Afilalo J, et al.. Guidelines for the echocardiographic assessment of the right heart in adults: A report from the American Society of Echocardiography. J Am Soc Echocardiogr
. 2010; 23 (7): 685–713.
26. Scharf M, Brem MH, Wilhelm M, Schoepf UJ, Uder M, Lell MM. Atrial and ventricular functional and structural adaptations of the heart in elite triathletes assessed with cardiac MR imaging. Radiology
. 2010; 257 (1): 71–79.
27. Scharhag J, Schneider G, Urhausen A, Rochette V, Kramann B, Kindermann W. Athlete’s heart
: right and left ventricular mass and function in male endurance athletes and untrained individuals determined by magnetic resonance imaging
. J Am Coll Cardiol
. 2002; 40 (10): 1856–1863.
28. Spence AL, Naylor LH, Carter HH, et al.. A prospective randomised longitudinal MRI study of left ventricular adaptation to endurance and resistance exercise training
in humans. J Physiol
. 2011; 589 (22): 5443–5452.
29. Teske AJ, Prakken NH, De Boeck BW, et al.. Echocardiographic tissue deformation imaging of right ventricular systolic function in endurance athletes. Eur Heart J
. 2008; 30 (8): 969–977.
30. Teske AJ, Prakken NH, De Boeck BWL, Velthuis BK, Doevendans PA, Cramer MJM. Effect of long term and intensive endurance training in athletes on the age related decline in left and right ventricular diastolic function as assessed by Doppler echocardiography. Am J Cardiol
. 2009; 104 (8): 1145–1151.
31. Vickers AJ, Altman DG. Statistics notes: analysing controlled trials with baseline and follow up measurements. BMJ
. 2001; 323 (7321): 1123–1124.
32. Wernstedt P, Sjostedt C, Ekman I, et al.. Adaptation of cardiac morphology and function to endurance and strength training—a comparative study using MR imaging and echocardiography in males and females. Scand J Med Sci Sports
. 2002; 12 (1): 17–25.