Athletic heart: a metabolic, anatomical, and functional study


Medicine & Science in Sports & Exercise:
Clinical Sciences: Clinical Investigations

Previous studies have suggested a reduced glucose uptake by the athlete's heart at rest. To examine whether there is a compensatory increase in the myocardial fatty acid utilization, we studied nine male endurance-trained athletes (age 26 ± 2 yr, ˙VO2max 60 ± 1 ml·kg-1·min-1, mean ± SEM) and eight sedentary subjects (age 26 ± 1 yr, ˙VO2max 38 ± 2 ml·kg-1·min-1) by single photon emission tomography using 123I-heptadecanoic acid (HDA) and mathematical modeling. Magnetic resonance imaging (MRI) and echocardiography were performed for the measurements of cardiac dimensions and left ventricular (LV) mass. No significant differences were found in the myocardial HDA beta-oxidation index(5.2 ± 2.0 vs 7.4 ± 1.6 μmol·min-1·100 g-1, P = NS) between endurance-trained and sedentary subjects. Fractional amounts of HDA beta-oxidation, backdiffusion, and esterification were also similar. In MRI study, LV mass was greater in the trained subjects (213 ± 9 vs 179 ± 10 g, P < 0.01) and in particular, LV long-axis diameter measured from the mitral valve level to the apex was increased (102 ± 2 vs 88 ± 2 mm, P< 0.001, trained vs sedentary subjects). ˙VO2max correlated with LV long-axis diameter (r = 0.77, P < 0.001). In contrast to our hypothesis, myocardial HDA utilization was not enhanced in endurance-trained athletes at rest. Increases in LV mass and especially in LV long-axis diameter were observed in the athletes, indicating LV longitudinal remodeling.

Author Information

University of Kuopio and Kuopio University Hospital, FIN-70211 Kuopio, FINLAND

Submitted for publication May 1995.

Accepted for publication August 1995.

This study was financially supported by Finnish Academy of Science and Finnish Ministry of Education.

Present address for Hannu Litmanen is Kuopio Research Institute of Exercise Medicine, Kuopio, and for Veikko A. Koivisto, Department of Medicine, University of Helsinki, Helsinki, Finland.

Address for correspondence: Anu Turpeinen, M.D., Department of Clinical Nutrition, University of Kuopio, P.O. Box 1627, FIN-70211 Kuopio, Finland.

Article Outline

Glucose, free fatty acids and lactate are the major myocardial fuels, and their proportion of the myocardial energy production varies a lot depending on the nutritional state, the substrate concentrations, myocardial work, oxygen demand and supply, and coronary flow (22). Under fasting conditions the most important fuel for myocardium is free fatty acids (FFA), which produce 60-70% of myocardial energy. Randle et al.(25) described the glucose-fatty acid cycle concerning the interaction between glucose and fatty acid metabolism in muscle and adipose tissue. The glucose-free fatty acid cycle operates also in the human heart, as shown in studies using positron emission tomography (PET)(21). Recently, it has been suggested that in endurance athletes myocardial glucose uptake is reduced as assessed by18 F-fluoro-2-deoxy-D-glucose (FDG) and PET (20). These data suggest either a reduced need for energy at rest or an alternative fuel, such as free fatty acids or lactate for the athlete's heart. By using single-photon emission tomography (SPET) and 123I-heptadecanoic acid(HDA), it is possible to evaluate myocardial fatty acid uptake noninvasively(15) and thus examine the possibility of increased use of free fatty acids in the athlete's heart at rest. Clearance of123 I-HDA in the myocardium is biexponential; the relative size and clearance half-time of the early phase have been used for indices of beta-oxidation, whereas the slow phase most likely reflects fatty acid turnover into cytosolic lipid pool (31). Deiodination of123 I-HDA and release of free radioiodine (123I-) has been shown to be related to fatty acid oxidation (16). The first purpose of the study was to examine whether the free fatty acid uptake is compensatorily increased in the endurance athlete's heart at rest.

The observation of correlation between myocardial glucose uptake and left ventricular (LV) mass and wall stress may suggest that myocardial hypertrophy and wall thickening contribute to the lowered glucose uptake in the athlete's heart (20). MRI is a modern and accurate imaging method in assessing cardiac volumes and mass (12), but only a few MRI studies have evaluated LV mass and dimensions in the athletes. Highly trained male (19) and female (26) endurance athletes had increased LV mass when studied by MRI. In another study young weight lifters showed increased wall thickness and diastolic internal dimensions of the left ventricle, whereas systolic LV internal dimension was decreased (8). Our second aim was to study cardiac dimensions, volumes, function, and mass in endurance athletes with magnetic resonance imaging (MRI) and M-mode and Doppler echocardiography to find out whether structural changes in the athlete's heart are linked with alterations in myocardial metabolism.

Back to Top | Article Outline



Nine endurance-trained athletes and eight sedentary subjects were studied. All of them were healthy males aged 19-34 yr and none were taking any medication. Five of the trained subjects were triathletes and four of them were competing at the national level. The other four athletes were cross-country skiers and one of them was a member of Finnish National ski team. All the athletes had been training regularly for years. The sedentary subjects did not exercise on a regular basis. All subjects gave their informed consent and the study was approved by the Ethical Committee of the University of Kuopio.

Back to Top | Article Outline

SPET study and echocardiography were carried out in the same morning between 0800 and 1200 h. Before the SPET study the subjects had fasted for at least 12 h and before fasting they had been on a diet containing carbohydrates at minimum 250 g·d-1 for at least 3 d. They were advised to avoid coffee, tobacco, and alcohol consumption and not to exercise for 2 d before the study. MRI, cardiopulmonary exercise test, and background laboratory examinations were performed within 2 wk of the SPET study.

Back to Top | Article Outline
Clinical Examination, Anthropometric Measurements, and Laboratory Examinations

All subjects were healthy as judged by clinical examination and standard laboratory tests. Standing height was measured without shoes to the nearest 0.5 cm and body weight by an electric weighing scale (Seca 708, Hamburg, Germany). Body mass index was calculated as weight/height squared(kg·m-2). The body composition was obtained by measuring skinfold thicknesses (biceps, triceps, subscapularis, suprailiac) with a Harpenden caliper (John Bull, British Indicators Ltd., St. Albans, Herts., UK); the skinfold thicknesses were summed and the fat mass was calculated according to Durnin and Womersley (4). Fasting blood glucose was analyzed by the glucose oxidase method (Glucose Auto & Stat, model GA-110, Daiici, Kyoto, Japan). Plasma insulin was analyzed by using Phadeseph Insulin RIA 100-kit (Pharmacia Diagnostics, Uppsala, Sweden). Serum free fatty acid concentration was measured by turbidometric method and analyzed with the specific-analyzer (Kone Ltd., Espoo, Finland). Serum lactate concentration was analyzed by the lactate oxidase method (YSI 2300 Stat, YSI Inc., U.S.A.).

Back to Top | Article Outline
Myocardial Free Fatty Acid Uptake

One hour before, and 12 and 24 h after 123I-HDA (Cygne bv, Eindhoven, The Netherlands) injection, 200-mg potassium perchlorate was given orally to block thyroid 123I-uptake. The patient was lying supine on the measuring table for 10-15 min before tracer injection. Heart rate and blood pressure were recorded and the double product was calculated to assess the myocardial work. The left antecubital vein was cannulated for collecting venous blood samples. A dose of 160-185 MBq of 123I-HDA was given intravenously into the right antecubital vein. The specific activity of the tracer was 15 × 1013 Bq·kg-1 and its radiochemical purity was better than 98%. Immediately after the injection of the tracer the first SPET scan was started using a dedicated three-headed gamma camera with high resolution collimators (Siemens MultiSPECT 3, Siemens Gammasonics Inc., Des Moines, IA). Full 360° rotation was used and thirty views per head(4°), each 18 s, were acquired. This was repeated six times consecutively. Thus, there were seven measurement points: 5, 15, 25, 35, 45, 55, and 65 min(as given in the mid time of each SPET scan). Venous blood samples of 5 ml were collected at 1.5, 3, 6, 10, 20, and 60 min after the injection of the tracer. A two-exponential curve fit was applied to the time activity curve of the blood samples, and it served as an input for the heart.

The 7-mm thick transaxial, sagittal, and coronal slices were reconstructed using a Butterworth filter (order of 8 and cut-off frequency 0.7 cm-1). Two consecutive coronal slices were added together and the “mid ventricular” slice (14 mm thick) was used for semiquantative analysis. Two circular regions of interest (ROI) were drawn, one over the myocardium and one over the middle of the left ventricular lumen. No spillover or overlapping compartment corrections were applied. The ratio of myocardium-to-lumen was calculated for each of the seven points. The average counts of the LV lumen were related to the counts of the extrapolated time activity curve of the blood samples. Thus, the assumption is that these two blood contents have to be equal radioactivity. Finally, two time activity curves were derived for further analysis; one for blood pool (input function) and one for myocardium(residue function).

Back to Top | Article Outline
Data Analysis

The modified FFA-model of Schelbert (29) was used and it is shown in Figure 1. The first compartment describes the plasma pool, the second one is the cell and the third one is mitochondrion. The fourth compartment describes the slow turnover pool of esterified 123I-HDA. Six rate constants between the compartments were used. The first one, k21, simply reflects myocardial perfusion and capillary permeability. This perfusion phase is followed by two elimination phases. The first elimination phase is considered to represent beta-oxidation, k32, and it is clinically the most important. The parameter k42 mainly reflects esterification of FFA-CoA into a slow turnover pool. The transfer rate constant k24 was fixed to zero by assuming that within the first 70 min no backflow from this slow turnover pool occurs, since DeGrado et al. (2) showed that k24 is 50 times slower than k42 in the rat heart. The parameter k12 describes backdiffusion of unused 123I-HDA into the vascular space and k13, its end-products which clear from the myocardium. Thus, there are five transfer constants (k24 = 0) that were iteratively adjusted to the plasma (input) and myocardium (residue) time activity curves using the SCOP/SCOPFIT program from the Duke University (14).

A mathematical index for beta-oxidation, MR(μmol·min-1·100 g-1 of myocardium) can be directly calculated from the model: eqn (1)

where s-FFA is serum fatty acid concentration (μmol·ml-1) and kij are the transfer rate constants of the model inFigure 1 (min-1). The percentage amounts used for beta-oxidation, for backdiffusion of non-esterified tracer and for esterification into the slow turnover pool were calculated as follows(2): eqns(2A),(2B),(2C)

These calculated estimates are not precisely equivalent to true myocardial fatty acid beta-oxidation, but rather functional parameters or so-called mathematical indices.

Back to Top | Article Outline
Magnetic Resonance Imaging

Seven endurance-trained and eight sedentary subjects underwent MRI study. MRI was performed at a magnetic field strength of 1.5 Tesla (Siemens Magnetom SP4000 Erlangen, Germany). ECG-triggered images were acquired in the true short- and long-axis planes of the heart, which were derived from axial, coronal, and sagittal localizer images. Short-axis cine MRI was performed using a gradient echo sequence (TR 50 ms, TE 12 ms, flip angle 60°). One slice was obtained through the level of papillary muscles. Correspondingly, long-axis cine was performed using the flip angle of 30° and one slice was obtained at the level of mitral valve to the apex. The acquisition matrix was 128×256 and the field of view 300 mm. Two acquisitions were averaged to improve signal-to-noise ratio. The number of frames per cardiac cycle was 16.

Long-axis cine images were analyzed by the software supplied by the manufacturer. Left ventricular and myocardial outlines were manually drawn on the end-diastolic and end-systolic images (Fig. 2A). This yielded left ventricular and myocardial dimensions at the end-diastole and end-systole. Left ventricular and myocardial volumes were approximated using the area-length method: Equation

where V is the ventricular/myocardial volume, A is the area and L is the length of the longitudinal axis. By using these volumes it was possible to estimate stroke volume and ejection fraction. Left ventricle wall thickness at the level of papillary muscles was measured from the short-axis cine images.

Left ventricle mass at the end-diastole was measured by using a fast gradient echo sequence (Turbo-Flash; TR 6.5 ms, TE 3 ms, flip angle 15°). Inversion time (TI) was 600 ms, which yielded the optimum contrast between the myocardium and blood. Matrix size was 64×128. Acquisition time for one data slice with this sequence was less than 1 s and up to 12 data slices 1 cm thick were acquired sequentially from the base of heart to the apex in the true short axis orientation. For each slice the epicardial and endocardial borders were manually traced and the epicardial and endocardial areas were calculated (Fig. 2B). The sum of all slices was taken as left ventricular and myocardial volume. Accordingly, the following formula was used to obtain LV mass: Equation

where Σ is summation over the slices involved, Aepi and Aendo are the areas enclosed by the epi- and endocardium (cm2) and thk is the slice thickness (cm). The volume was expressed in cm3(ml).

Aortic flow velocity was measured using phase-mapping technique(5). Parameters for phase-mapping were TR 34 ms, TE 6 ms, and flip angle 20°. For reconstruction of time-velocity curve, 32 single velocity values were extracted from consecutive heart cycles. The time interval between the two consecutive velocity values depended on individual heart rate and ranged between 32 and 50 ms. Data acquisition required about 5 min to obtain a two-dimensional space-resolved and time-averaged blood flow profile. Measuring plane was perpendicular to the aortic root. For off-line measurement of the peak systolic velocity, a ROI of one pixel was placed in the middle of the aorta to allow optimal correlation with the Doppler-derived data.

Back to Top | Article Outline
M-Mode and Doppler Echocardiography

Echocardiographic measurements were performed in the left lateral semirecumbent position as previously described (33). M-mode and Doppler tracings (100 mm·s-1 paper speed) were obtained with a simultaneous electrocardiogram with Aloka SSD-870 ultrasound system using either 2.5 MHz or 5 MHz transducer (Aloka Ltd., Tokyo, Japan). All tracings were analyzed by using a digitizing table (MM 1201 Summagraphics Co, Fairfield, CT; resolution 0.1 mm) and the measurements were obtained from 3 to 5 cardiac cycles and averaged. M-mode measurements were performed according to the recommendations of the American Society of Echocardiography using the leading edge technique (27). Doppler signals were recorded at the level of mitral leaflet tips and left ventricular diastolic function indexes were calculated (33). Left ventricular mass was calculated according to Devereux and Reichek(3): LV mass = 1.04[(LVEDD3 + IVS3 + PW3) - LVEDD3] - 13.6 g. LV wall stress(10) and end-diastolic and end-systolic volumes(24) were calculated as previously described.

Back to Top | Article Outline
Cardiopulmonary Exercise Test

Maximum aerobic capacity was evaluated in an incremental bicycle ergometer(Ergoline 900) test to the limit of tolerance. After 2 min of unloaded exercise the work rate was increased by 35 W·min-1. Continuous respiratory gas exchange measurements were taken breath-by-breath(SensorMedics MMC 4400, Irvine, CA) and averaged for 30 s. Maximum oxygen uptake (˙VO2max) (34) was obtained as a highest 30-s averaged value. True maximal oxygen uptake values (plateauing of˙VO2 at the end of the test with increase <150 ml·min-1) were obtained in seven of nine subjects in the athlete group and in four of eight subjects in the sedentary group. However, according to Cooper et al. (1) maximum ˙VO2 obtained from incremental testing with legs closely approximates the predicted maximal oxygen uptake even when the plateau is not reached. During exercise, heart rate was measured from the electrocardiogram and blood pressure by using a sphygmomanometer and stethoscope.

Back to Top | Article Outline
Statistical Methods

The differences between the endurance-trained and sedentary subjects were analyzed by Student's unpaired t-test. Analyses of variance and covariance was used in analyzing cardiac diastolic function (heart rate as a co-variant). Due to large variation, myocardial HDA beta-oxidation index, and percentages of beta-oxidation, backdiffusion and esterification into the slow turnover pool were log-transformed. Spearman's rank correlations were used where appropriate. Differences were considered to be statistically significant if the P-value was ≤0.05 (two-tailed). All results are expressed as the mean ± SEM. Data were analyzed by SPSS/PC statistical package(version 4.01, SPSS Inc., Chicago, IL).

Back to Top | Article Outline


Characteristics of the Subjects

The endurance-trained and sedentary subjects were comparable in age, weight and body mass index, but the former group had a lower mean percent body fat(Table 1). Systolic blood pressure was equal in the two groups, whereas heart rate was lower in the trained subjects, and thus the double product during the SPET study was smaller. Maximum oxygen uptake was higher in trained than in sedentary subjects (Table 1), whereas blood pressure (205 ± 11 vs 201 ± 7 mm Hg, trained vs sedentary subjects) and heart rate (185 ± 3 vs 190 ± 4 beats·min-1) at the maximal level of exercise did not differ between the groups.

Back to Top | Article Outline
Myocardial Fatty Acid Uptake

Serum FFA concentration was similar in the two groups (0.49 ± 0.11 vs 0.51 ± 0.07 mmol·l-1, trained vs sedentary subjects, NS). Fasting blood glucose and lactate concentrations were also comparable(glucose 4.0 ± 0.1 vs 4.4 ± 0.1 mmol·l-1, lactate 0.76 ± 0.06 vs 0.84 ± 0.12 mmol·l-1, trained vs sedentary subjects, both NS), but fasting plasma insulin concentration was slightly lower in the trained subjects (44 ± 5 vs 68 ± 6 pmol·l-1, P < 0.05). Initial uptake of123 I-HDA did not differ significantly between the two groups (4.0± 0.5 vs 5.1 ± 0.4% of injected dose·100 g-1, trained vs untrained subjects, NS). Myocardial FFA beta-oxidation index (5.2± 2.0 vs 7.4 ± 1.6 μmol·min-1·100 g-1) normalized for tissue weight was similar in endurance-trained and sedentary subjects, as was also total myocardial FFA beta-oxidation (14.8± 6.0 vs 13.9 ± 2.6 μmol·min-1, NS).

In endurance-trained subjects, transfer rate constant k32 for beta-oxidation was only around a half of that in sedentary subjects (0.053± 0.010 vs 0.101 ± 0.031). However, this difference was not statistically significant (P = 0.17). Nor did other transfer rate constants differ between the two groups (data not shown). When myocardial HDA utilization was further divided into percentages of beta-oxidation, backdiffusion into the vascular space and esterification into the slow turnover pool, beta-oxidation accounted for 45% and backdiffusion for 49% in the athletes (vs 55% and 39% in the sedentary subjects). These differences were not statistically significant. Esterification into the slow turnover pool was 6% in both groups.

There was a large individual variation in the myocardial FFA beta-oxidation index (range 0.5-19.6 μmol·min-1·100 g-1) and percentages of beta-oxidation, backdiffusion, and esterification into the slow turnover pool, but the variation was not attributable to the training, maximal aerobic capacity, cardiac dimensions or any other variable.

Back to Top | Article Outline
Cardiac Dimensions and Function

When studied with MRI, the endurance-trained subjects had increased LV mass, end-diastolic volume and stroke volume compared with sedentary subjects(Table 2). In particular, LV long-axis diameter measured in end-diastole from the mitral valve level to the apex was significantly greater in the trained subjects, and also posterior wall thickness was increased compared with the sedentary subjects (Table 2). LV short-axis end-diastolic diameter and end-systolic volume as well as anterior, septal, and inferior wall thicknesses tended to be greater in the trained subjects, but these differences were not statistically significant. Ejection fraction and peak aortic flow velocity did not differ between the two groups (Table 2).

In the echocardiographic study especially LV mass index and end-diastolic diameter but also LV end-systolic diameter and interventricular septal thickness were significantly greater in the endurance-trained subjects when compared with sedentary subjects (Table 3). LV mass was 288 ± 14 g in the trained group and 201 ± 13 g in the sedentary group (P = 0.001). LV end-diastolic volume (161 ± 7 vs 113± 10 ml, P < 0.001) and endsystolic volume (42 ± 4 vs 32 ± 3 ml P < 0.05) were also increased in the trained subjects. Concerning LV diastolic function, the endurance-trained subjects had a higher LV peak early filling velocity integral and longer deceleration half-time of early peak flow velocity (Table 3). After adjustment for heart rate, the difference between the groups in deceleration half-time still persisted, whereas LV peak early filling velocity integral was no longer significantly different between the groups. LV peak early filling velocity/peak atrial filling velocity (E/A) ratio tended to be greater in the trained subjects (P = 0.08). LV wall stress as well as systolic function assessed by fractional shortening were similar in the trained and sedentary subjects (Table 3).

Back to Top | Article Outline

˙VO2max had a positive correlation with MRI LV long-axis diameter (r = 0.77, P < 0.001), MRI LV end-diastolic volume (r = 0.75, P < 0.001), and stroke volume (r = 0.78, P < 0.001), but also with echocardiographical measurements (with LV mass index r = 0.68, P < 0.001 and with LV end-diastolic diameter r = 0.59,P < 0.01, respectively). LV masses obtained from MRI and echocardiography only moderately correlated with each other (r = 0.47,P = 0.05), as did also interventricular septum (r = 0.44,P = 0.06) and posterior wall thicknesses (r = 0.46, P = 0.05). A stronger correlation was found in measurements of LV volumes (LV end-diastolic volume r = 0.71, P < 0.001 and endsystolic volume r= 0.61, P < 0.01).

Transfer constant rate k32 had an inverse correlation with MRI anterior wall thickness (r = -0.65, P < 0.01), and the percentage of beta-oxidation in the myocardium also had an inverse correlation with MRI stroke volume (r = -0.65, P < 0.01). Otherwise we did not observe any significant linear correlations between myocardial fatty acid uptake and cardiac dimensions or function. Nor did LV wall stress or double product associate with fatty acid utilization.

Back to Top | Article Outline


Exercise training causes a number of well-known physiologic changes in the heart: increases in LV end-diastolic diameter and wall thicknesses lead to increased LV mass (17), stroke volume is increased, and heart rate is decreased during resting conditions. To what extent these adaptative structural and functional changes are linked with metabolic changes in the myocardium is not well known. During the last years the development of modern imaging methods PET and SPET has made it possible to examine myocardial metabolism in vivo. Nuutila and coworkers (20) have recently observed that in endurance athletes myocardial glucose uptake is reduced at rest compared with sedentary subjects, in contrast to increased glucose uptake in the skeletal muscle. They concluded that this could be due to either lower energy requirement in the myocardium or the use of alternative fuels, such as free fatty acids or lactate. By using radioiodinated fatty acids, such as 123I-HDA, it is possible to evaluate myocardial free fatty acid metabolism with SPET (15). 123I-HDA is taken up in the normal myocardium and oxidized with concomitant release of the radioiodine (16,30).

Our aim was to examine the possibility of increased use of an alternative fuel, free fatty acids, in the athlete's heart under standardized conditions. SPET study was performed at rest in the morning after an overnight fast to reach maximal serum FFA concentration and the subjects were told not to exercise for 2 d before the SPET study. Despite standardized conditions, there was a large inter-individual variation in the myocardial HDA beta-oxidation index and percentages of beta-oxidation, backdiffusion, and esterification into the slow turnover pool within the endurance-trained and sedentary groups. The variation was not related to endurance training, maximal exercise capacity, heart size, or any known other variable; rather, it was random. In myocardial FFA utilization, no significant differences were observed between endurance-trained and sedentary subjects. Thus, the results show that myocardial FFA utilization is not increased in endurance-trained athletes. Taken together with the findings of Nuutila et al. (20), it seems that reduced glucose uptake by the athlete's heart is not accompanied by a compensatory increase in FFA utilization, and it is possible that in endurance-trained athletes myocardial energy requirement may be lowered on the whole at rest. It has to be emphasized that SPET imaging with 123I-HDA only indirectly estimates myocardial free fatty acid utilization, but on the other hand, direct measurements demanding highly invasive procedures would not have been possible in this study. Moreover, with the present SPET imaging techniques it is not possible to study simultaneously myocardial glucose and FFA uptake, which would be of interest.

There are some differences in the experimental design and the training status of the athletes between this study and Nuutila et al. First, Nuutila et al. (20) reported smaller LV wall stress in the athletes and it correlated with myocardial glucose uptake, suggesting that a reduced wall stress may result in lowered myocardial energy requirement in the athlete heart. In contrast, we found the LV wall stress to be exactly the same in trained and sedentary subjects. Previously, lowered wall stress in endurance athletes at rest has been reported (23), but the results are controversial (28). One can also question whether the endurance-trained subjects in our study were highly conditioned athletes, since their mean ˙VO2max determined in this study appeared to be about 60 ± 1 ml·kg·min-1 (vs 72 ± 2 ml·kg-1·min-1 in the Nuutila's study). However, the maximal exercise capacity of these national class endurance athletes was determined by bicycle ergometer instead of treadmill, and it has been observed that the latter gives approximately 5-10% higher maximum oxygen uptake values than the former (11). It is obvious that by using a treadmill higher maximal oxygen uptake values could have been achieved. Third, while Nuutila et al. measured glucose uptake during hyperinsulinemic clamp, in the current study fatty acid uptake was determined in the fasting state with low plasma insulin concentrations. In this study the endurance-trained subjects had lower fasting plasma insulin concentration than the sedentary subjects, reflecting enhanced insulin sensitivity of trained subjects(13). Thus, these two studies are not totally comparable regarding the design or the subjects. This obviously influences the interpretation of the results.

Concerning MRI results, it is noteworthy that it was the LV long-axis diameter measured from the mitral valve level to the apex that was most significantly increased in the endurance-trained subjects compared with sedentary subjects. Previous MRI studies performed in endurance athletes(19,26) or weightlifters (8) have not observed this. In numerous echocardiographic studies performed in endurance athletes, usually only an increase in LV transversal end-diastolic diameter has been reported; and although in practice LV long-axis lengthening is frequently seen, only little attention has been paid to it. Use of MRI could be of valuable help in reporting accurate LV short-axis and long-axis dimensions. Not only was LV volume increased in the athletes, but also LV wall-thickening contributed to the increased cardiac mass.

In the echocardiographic study the results concerning LVEDD, LV mass, and wall thicknesses were very similar compared with previous observations reviewed by Maron (17) and Urhausen and Kindermann(32). Despite increased cardiac mass, in most of the studies diastolic function have been observed to remain normal in the trained heart at rest (6,7). At high heart rates, Gledhill et al. (9) recently reported enhancement in LV filling in the athletes; also LV emptying was enhanced compared with sedentary subjects. Matsuda et al. (18) observed that early diastolic filling during exercise was better in the athletes. We found the peak early LV filling velocity integral and deceleration half-time to be greater in athletes at rest. However, lower heart rate may have contributed to this finding to some extent.

In conclusion, myocardial FFA uptake is not enhanced in the athlete's heart at rest. Increases in LV mass, long-axis diameter and volumes, but also in posterior wall thickness in endurance-trained athletes can reliably be observed by MRI.

Back to Top | Article Outline


1. Cooper, D. M., D. Weiler-Ravell, B. J. Whipp, and K. Wasserman. Aerobic parameters of exercise as a function of body size during growth in children. J. Appl. Physiol. 56:628-635, 1984.
2. Degrado, T. R, J. E. Holden, C. K. Ng, D. M. Raffel, and S. J. Gatley. Quantative analysis of myocardial kinetics of 15-p-[Iodine-125)Iodophenylpentadecanoic acid. J. Nucl. Med. 30:1211-1218, 1989.
3. Devereux, R. B. and N. Reichek. Echocardiographic determination of left ventricular mass in man. Anatomic validation of the method. Circulation 55:613-618, 1977.
4. Durnin, J. V. G. A. and J. Womersley. Body fat assessed from total body density and its estimation from skinfold thicknesses: measurements on 481 men and women aged 16 to 72 years. Br. J. Nutr. 32:77-92, 1974.
5. Engels, G., E. Muller, K. Reynen, N. Wilke, and K. Bachmann. Phase-mapping technique for the evaluation of aortic valve stenosis by MR. Eur Radiol. 2:229-304, 1992.
6. Fagard, R., C. Van Den Broeke, L. Vanhees, J. Staessen and A. Amery. Noninvasive assessment of systolic and diastolic left ventricular function in female runners. Eur. Heart J. 8:1305-1311, 1987.
7. Finkelhor, R. S., L. J. Hanak, and R. C. Bahler. Left ventricular filling in endurance-trained subjects. J. Am. Coll. Cardiol 8:289-293, 1986.
8. Fleck, S. J., C. Henke, and W. Wilson. Cardiac MRI of elite junior olympic weight lifters. Int. J. Sports Med. 10:329-333, 1989.
9. Gledhill, N., D. Cox, and R. Jamnik. Endurance athletes' stroke volume does not plateau: major advantage is diastolic function.Med. Sci. Sports Exerc. 26:1116-1121, 1994.
10. Grossman, W., D. Jones, and L. P. McLaurin. Wall stress and patterns of hypertrophy in the human left ventricle. J. Clin. Invest. 56:56-64, 1975.
11. Hermansen, L. and B. Saltin. Oxygen uptake during maximal treadmill and bicycle exercise. J. Appl. Physiol. 26:31-37, 1969.
12. Katz, J., M. C. Milliken, J. Stray-Gundersen et al. Estimation of human myocardial mass with MRI imaging. Radiology 169:495-498, 1988.
13. Koivisto, V. A., H. Yki-Järvinen and R. A. DeFronzo. Physical training and insulin sensitivity. Diab. Metab. Rev. 1:445-481, 1986.
14. Kootsey, J. M. Introduction to computer simulation. National Simulation Resource, Duke University, Durham, NC, with support from NIH grant RR01693.
15. Kuikka, J. T., J. N. Mustonen, M. I. J. Uusitupa, et al. Demonstration of disturbed free fatty acid metabolism of myocardium in patients with non-insulin dependent diabetes mellitus as measured with iodine-123-heptadecanoic acid. Eur. J. Nucl. Med. 18:475-481, 1991.
16. Lüthy, P., P. Chatelain, I. Papageorgiou, A. Schubiger, and R. A. Lerch. Assessment of myocardial metabolism with iodine-123 heptadecanoic acid: effect of decreased fatty acid oxidation on deiodination. J. Nucl. Med. 29:1088-1095, 1988.
17. Maron, B. J. Structural features of the athlete heart as defined by echocardiography. J. Am. Coll. Cardiol. 7:190-203, 1986.
18. Matsuda, M., Y. Sugishita, S. Koseki, I. Ito, T. Akatsuka, and K. Takamatsu. Effect of exercise on left ventricular diastolic filling in athletes and nonathletes. J. Appl. Physiol. 55:323-328, 1983.
19. Milliken, M. C., J. Stray-Gundersen, R. M. Peshock, J. Katz, and J. H. Mitchell. Left ventricular mass as determined by magnetic resonance imaging in male endurance athletes. Am. J. Cardiol 62:301-305, 1988.
20. Nuutila, P., M. J. Knuuti, O. J. Heinonen, et al. Different alterations in the insulin-stimulated glucose uptake in the athlete's heart and skeletal muscle. J. Clin. Invest. 93:2267-2274, 1994.
21. Nuutila, P., V. A. Koivisto, J. Knuuti, et al. The glucose-free fatty acid cycle operates in human heart and skeletal musclein vivo. J. Clin. Invest. 89:1767-1744, 1992.
22. Opie, L. H. Fuels: carbohydrates and lipids. In:The Heart: Physiology and Metabolism. New York: Raven Press, 1991, pp. 208-246.
23. Percy, R. F., D. A. Conetta, and A. B. Miller. Echocardiographic assessment of the left ventricle of endurance athletes just before and after exercise. Am. J. Cardiol. 65:1140-1144, 1990.
24. Pombo, J. F., B. L. Troy, and R. O. Russell, Jr. Left ventricular volumes and ejection fraction by echocardiography.Circulation 43:480, 1971.
25. Randle, P. J., P. B. Garland, C. N. Hales, and E. A. Newsholme. The glucose-fatty acid cycle. Its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet 1:785-789, 1963.
26. Riley-Hagan, M., R. M. Peshock, J. Stray-Gundersen, J. Katz, T. W. Ryschon, and J. H. Mitchell. Left ventricular dimensions and mass using magnetic resonance imaging in female endurance athletes. Am. J. Cardiol 69:1067-1074, 1992.
27. Sahn, D. J., A. Demaria, J. Kisslo, and A. Weyman. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurement. Circulation 58:1072-1083, 1978.
28. Schairer, J. R., S. Keteyian, J. W. Henry, and P. D. Stein. Left ventricular wall tension and stress during exercise in athletes and sedentary men. Am. J. Cardiol. 71:1095-1098, 1993.
29. Schelbert, H. R. Features of positron emission tomography as a probe for myocardial chemistry. Eur. J. Nucl. Med. 12:S2-S10, 1986.
30. Sloof, G. W., F. C. Visser, M. J. Van Eenige, et al. Comparison of uptake, oxidation and lipid distribution of 17-iodoheptadecanoic acid, 15-(p-iodophenyl)pentadecanoic acid and 15-(p-iodophenyl)-3,3-dimethylpentadecanoic acid in normal canine myocardium.J. Nucl. Med. 34:649-657, 1993.
31. Syrota, A. and P. Jehenson. Complementarity of magnetic resonance spectroscopy, positron emission tomography and single photon emission tomography for the in vivo investigation of human cardiac metabolism and neurotransmission. Eur. J. Nucl. Med. 18:897-923, 1991.
32. Urhausen, A. and W. Kindermann. Echocardiographic findings in strength- and endurance-trained athletes. Sports Med. 13:270-284, 1992.
33. Vanninen, E., J. Mustonen, P. Vainio, E. Lansimies, and M. Uusitupa. Left ventricular function and dimensions in newly diagnosed non-insulin dependent diabetes mellitus. Am. J. Cardiol 70:371-378, 1992.
34. Wasserman, K., J. E. Hansen, D. Y. Sue, and B. J. Whipp. Principles of Exercise Testing and Interpretation. Philadelphia: Lea & Febiger, 1987, pp. 29-30.


©1996The American College of Sports Medicine