Seven highly trained competitive child cyclists (mean age 11.9 yr, range 10.7–13.2) were recruited for exercise testing. Eleven boys from the New England region were initially identified from the registration list of the United States Cycling Federation as potential subjects. Two declined to participate, one was not actively training, and one could not be located. The cyclists in the study were all healthy, and none was taking medication that would influence aerobic fitness. Average duration of training was 3.4 yr (range 1.5–6 yr). Cyclists had been training an average of 4 d·wk−1 (range 2–7 d). One was a triathlete who spent most of his time training in distance running, whereas another competed principally in bicycle moto-cross (BMX) racing. The other five cycled for an average of 83 miles·wk−1 (range 45–125 miles). All subjects were competing in regional and national races at least weekly at the time of testing. All cyclists were participants in other sports (soccer, swimming, basketball, or rock climbing).
To determine pubertal status, subjects were asked to self-assess appearance of secondary sexual characteristics. None had developed facial hair or voice change, although two of the seven reported appearance of pubic hair.
Findings in the cyclists were compared with those of 39 nontraining boys, who have previously been described in reports of the influence of cardiovascular fitness on distance run performance (28) and the physiological determinants of maximal oxygen uptake (27). To assure a wide range of fitness, this group was comprised of 10 boys recruited from each quartile of finish times in a mile run test (one failed to complete testing). Twenty-seven (69%) had been participating on a community sports team (mainly soccer and basketball), but none had been involved in regular exercise training.
The untrained subjects were instructed to avoid intensive physical activity for the day before testing and avoid food intake within 2–3 h of testing. The cyclists were asked to prepare for the test (diet, sleep, and exercise) in the same way they would for a competitive race. Testing was performed in an air-conditioned laboratory with moderate humidity and temperature of 20°C.
Height was measured with a stadiometer and body mass (without shoes) with a calibrated balance beam scale. Scapular and triceps skin-fold measurements were taken in triplicate using standard techniques. Summed measurements were converted to percent body fat using the Slaughter equations (37). Lean body mass was calculated as (body mass) − (percent fat × body mass × 0.01).
Left ventricular dimensions were measured by M-mode echocardiography using two dimensional guidance (Hewlett Packard Sonos 1000, Andover, MA) with subjects in the supine position. Left ventricular measurements were made in the parasternal long axis view immediately distal to the tips of the mitral valve leaflets. The left ventricular end-diastolic dimension (EDD) was recorded as the distance from the posterior edge of the ventricular septum to the endocardial surface of the left ventricular free wall coincident with the Q wave of the electrocardiogram. The left ventricular end-systolic dimension (ESD) was measured as the shortest distance from the free wall endocardial surface to the ventricular septum during ventricular systole. Percent left ventricular shortening fraction (SF) was then calculated as the quotient (EDD − ESD)/EDD × 100. Left ventricular septal and posterior wall dimension in diastole were measured at the Q wave of the electrocardiogram. All dimensions were averaged from three measurements and were expressed relative to the square root of the body surface area (9).
Upright maximal cycle testing was performed with a mechanically braked Monark ergometer (Varberg, Sweden). Seat height was adjusted to produce an approximate 15° knee angle at full extension. Cyclists pedaled with an initial cadence of 50 rpm for 3-min workloads of 25 and 75 W. Pedal cadence was then increased to 90 rpm, with incremental loads of 25 W to the point of exhaustion. Nonathletes cycled with initial and incremental workloads of 25 W, with 3-min stages and a constant 50-rpm pedaling cadence. Exhaustion was defined as the point when the subject was no longer capable of maintaining the appropriate pedaling rate.
Heart rate was determined by electrocardiography. Gas exchange variables were measured with standard open circuit techniques using a Q-Plex Cardio-Pulmonary Exercise System (Quinton Instrument Company, Seattle, WA). Subjects breathed through a 94-mL dead-space Rudolph valve, and minute ventilation was measured by a pneumotachometer in the expiratory line. Expired air samples were drawn from a 6-L mixing chamber and analyzed for oxygen and carbon dioxide content by zirconia oxide and infrared analyzers, respectively. Mean values for VE, V̇O2, V̇CO2, and V̇CO2/V̇O2 (RER) were calculated over 15-s intervals. The system was calibrated before each individual test using standard gases of known oxygen and carbon dioxide concentration.
Peak oxygen uptake was determined as the mean of the two highest values (15-s averages) during the final min of exercise. Peak V̇O2 was considered equivalent to V̇O2max if subjective signs of exhaustion (hyperpnea or sweating) were evident, peak heart rate was over 185 bpm, and maximal RER exceeded 1.00 (22). Physical work capacity (PWC) was defined as the greatest workload achieved in W, prorated for partial workloads completed.
Standard Doppler echocardiographic techniques were used to estimate cardiac output (36). Velocity of blood in the ascending aorta was measured with a 1.9-MHz continuous Doppler transducer directed from the suprasternal notch. The integral of velocity over time (velocity-time integral, VTI) for individual beats was determined by tracing the contour of the velocity curve both on-line and off-line. VTI values at rest, during the final min of each 3-min workload, and in the final 30 s of exercise were obtained by averaging the 3–10 curves with the greatest VTI values and most distinct spectral envelopes.
The maximal diameter of the ascending aorta at the sinotubular junction was measured by two-dimensional echocardiography with the subject seated on the cycle ergometer immediately before exercise. The aortic cross-sectional area was then calculated from the mean of 5–10 diameter measurements, assuming the aorta to be circular.
Stroke volume was estimated as the product of VTI and aortic cross-sectional area. Cardiac output was then calculated as the product of heart rate and stroke volume, and arteriovenous oxygen difference was calculated as the quotient of V̇O2 divided by cardiac output. Cardiac output and stroke volume were expressed relative to body surface area (cardiac index and stroke index, respectively).
Although the Doppler technique is subject to potential methodological error (36), reliability and validity have been high. Among 13 subjects in this laboratory, the intraclass correlation coefficient for Doppler-derived maximal stroke volume during two identical cycling tests was R = 0.90, with a coefficient of variation of 8.5% (29). No significant difference in mean maximal stroke volume were observed when comparing Doppler measures with those determined by thoracic bioimpedance in eight boys (60 ± 11 mL vs 68 ± 13 mL) (30).
Systolic ejection time at maximal exercise was determined from the Doppler-derived velocity-time curves. Diastolic filling time at exhaustion was then estimated as cardiac cycle time (60/maximal heart rate) minus systolic ejection time (ignoring minor isovolumic systolic and diastolic periods). The systolic ejection velocity and diastolic filling velocity at maximal exercise were determined by dividing maximal stroke volume by systolic ejection and estimated diastolic filling periods, respectively, assuming that stroke volume corresponds to left ventricular filling volume per beat.
Physiological and anthropometric differences between cyclists and untrained subjects were examined by independent Student t-test. Statistical significance was defined as P < 0.05.
Informed consent and assent were obtained from the parents and children, respectively. This study was reviewed and approved by the Institutional Review Board of the Baystate Medical Center.
The two groups were similar in age (11.9 ± 0.8 yr for cyclists and 12.2 ± 0.5 yr for nonathletes). The cyclists demonstrated lower body fat than the untrained subjects (13.4 ± 3.2% vs 21.2 ± 9.1%, respectively) and also had a significantly lower weight (38.0 ± 4.1 vs 45.6 ± 10.1 kg), height (146 ± 6 vs 153 ± 9 cm), and body surface area (1.25 ± 0.09 vs 1.40 ± 0.18 m2).
Three of the nonathletes and two of the cyclists did not satisfy maximal heart rate criteria for a maximal effort. Based on subjective observation, however, it was considered that a true exhaustive test had been achieved, and the data on these subjects are included. Values for maximal stroke volume were obtained on all subjects. No significant difference was observed in maximal RER between the two groups (1.07 ± 0.04 for the cyclists and 1.06 ± 0.04 for the untrained boys). Average V̇O2max was 60.0 ± 6.0 and 47.0 ± 5.8 mL·kg−1·min−1 in the cyclists and nonathletes, respectively (P < 0.05). This difference persisted when values were expressed relative to lean body mass (69.4 ± 6.9 vs 58.2 ± 5.2 mL·kg LBM−1·min−1, respectively). Mean PWC, expressed relative to total body mass, was 4.02 ± 0.32 W·kg−1 in the cyclists and 3.04 ± 0.45 W·kg−1 in the nonathletes (P < 0.05).
Preexercise and maximal values for heart rate, stroke volume, cardiac output, and arteriovenous oxygen difference are outlined in Table 1. Both resting and maximal heart values were significantly lower in the cyclists. It should be noted, however, that two unusually low maximal rates in the cyclists (both 165 bpm) influenced the average maximal value in this group.
Mean preexercise values for cardiac index were almost identical in the cyclists and nonathletes (3.93 ± 0.66 vs 3.94 ± 0.83 L·min−1·m−2, respectively). Consequently, the lower resting heart rate in the cyclists was accompanied by a greater mean preexercise stroke index (59 ± 6 vs 44 ± 9 mL·m−2).
At the two submaximal stages at which the cyclists and nonathletes exercised at identical workloads and pedaling cadence (25 and 75 W, 50 rpm), no significant differences were observed in V̇O2 or cardiac output (Table 2). However, among the cyclists, the contribution of stroke volume was significantly greater and heart rate lower than those of the nonathletes.
At maximal exercise, cardiac index was significantly greater in the cyclists (13.94 ± 1.37 vs 11.95 ± 2.28 L·min−1·m−2), reflecting a higher maximal stroke index (76 ± 6 mL·m−2 vs 60 ± 11 mL·m−2) compared with the untrained subjects. No significant differences were observed between the two groups for maximal arteriovenous oxygen difference (13.1 ± 0.8 and 13.0 ± 2.5 mL·100 mL−1 for cyclists and nonathletes, respectively). Average peak aortic velocity at maximal exercise was 219 ± 37 cm·s−1 in the cyclists and 192 ± 38 cm·s−1 in the nonathletes (P > 0.05).
With the onset of exercise, the stroke volume rose in both groups, but beyond moderate exercise intensities no appreciable change was observed in either group (Fig. 1). In the cyclists, stroke index at an average exercise intensity of 54% V̇O2max (75 W) was 99% of the maximal stroke index. Among the nonathletes, the stroke index at a workload intensity equal to 46% V̇O2max (50 W) was 93% of the stroke index at exhaustion. No significant difference was seen in the mean ratio of maximal to resting stroke volume in the two groups (1.31 ± 0.11 for cyclists and 1.41 ± for nonathletes, P > 0.05).
The systolic ejection period at maximal exercise was similar in the cyclists (0.195 ± 0.019 s) and nonathletes (0.193 ± 0.015 s). Because cardiac cycle length was greater at exhaustion in the cyclists, diastolic filling period was significantly longer as well (0.132 ± 0.020 vs 0.110 ± 0.015 s). The larger maximal stroke volume in the cyclists resulted in a higher average rate of systolic ejection at maximal exercise (491 vs 435 cc·s−1), but the mean diastolic filling rate at peak exercise was greater in the nonathletes (763 vs 728 cc·s−1).
Average resting supine left ventricular EDD (indexed to the square root of body surface area) was 3.81 ± 0.16 cm·BSA−0.5 for the cyclists and 3.64 ± 0.31 cm·BSA−0.5 for the untrained boys (P > 0.05) (Fig. 2). No significant difference was observed in left ventricular diastolic septal thickness (6.4 ± 1.1 mm·BSA−0.5 for cyclists, 6.3 ± 1.3 mm·BSA−0.5 for nonathletes) or posterior wall thicknesses (5.7 ± 1.1 mm·BSA−0.5 and 5.7 ± 1.0 mm·BSA−0.5, respectively). The left ventricular shortening fraction was also similar in the two groups (36.4 ± 4.6% for cyclists and 39.1 ± 5.7% for nonathletes, P > 0.05).
The physiological profile of the highly trained child cyclists in this study resembles that previously described in adult endurance athletes (1,6,14,17,18,33). The cyclists demonstrated a higher maximal oxygen uptake, cardiac index, and stroke index compared with the untrained subjects, and at rest and a given submaximal workload their heart rates were lower and stroke volumes greater. Average resting left ventricular size was larger, although still in the normal range. Nonetheless, certain quantitative differences were observed from those reported in adult athletes.
The mean V̇O2max in the cyclists (60.0 mL·kg−1·min−1) in this study is consistent with values previously described in highly trained child endurance athletes. Reports of V̇O2max in prepubertal male distance runners have ranged from 59.3 to 66.5 mL·kg−1·min−1, approximately 20% greater than an expected average value in nontrained boys during treadmill running of 52 mL·kg−1·min−1 (21). In the present study, average V̇O2max was 27.6% higher in the cyclists than nonathletic subjects.
These values are lower than findings observed in adult endurance athletes. In the compilation by Faria et al. (7) of V̇O2max in 14 studies of adult elite cyclists, the average value was 71.7 mL·kg−1·min−1 with a range of 67.6–77.4 mL·kg−1·min−1. This represents a maximal aerobic power which is approximately 60% greater than that expected in the average young adult male.
This superior level of aerobic fitness in trained adult versus child athletes has been cited as evidence for a blunted plasticity of V̇O2max in response to endurance training in the prepubertal years (21). This concept is supported by a large volume of research data indicating that the magnitude of increase in V̇O2max after aerobic training of nonathletic children (usually 5–10%) is far less than that typically seen when young adults undergo similar training regimens (15–30%) (23).
Although hormonal influences at puberty might trigger a greater degree of aerobic trainability (24), other explanations might account for the greater V̇O2max values in highly trained adult endurance athletes. Reports in adults involve elite cyclists who are members of national and Olympic teams, a select group of athletes whose extremely high aerobic fitness reflects both intensive training and genetic endowment. Those adult athletes with lesser values of V̇O2max are not members of these teams and are not included in these reports. Such a “filtering” of those with particularly high aerobic fitness for elite competition does not occur at the pediatric level. This explanation is suggested by the observation that two of the boys in the present study had V̇O2max values of 69.9 and 67.4 mL·kg−1·min−1, well within the range observed among elite adult cyclists. Alternatively, it is possible that the development of high levels of V̇O2max simply requires more years and greater intensity of cycle training than can be accumulated by a 12-yr-old boy.
Both anthropometric and physiological factors were responsible for the greater values of V̇O2max (expressed per kg body mass) in the child cyclists in this study compared with nonathletic subjects. The cyclists exhibited significantly lower body fat content, an important determinant of V̇O2max·kg−1 (28). However, greater aerobic fitness persisted when V̇O2max was expressed relative to lean body mass (69.4 vs 58.6 mL·kg−1·min−1 in cyclists and nonathletes, respectively), indicating that true physiological differences separated the two groups.
Among the candidates offered by the Fick equation, maximal stroke volume was the sole physiological factor accounting for the greater V̇O2max values in the cyclists compared to nonathletic boys in this study. Mean stroke index at maximal exercise was 26.7% greater in the cyclists (76 vs 60 mL·m−2) and was responsible for their higher maximal cardiac index (mean 13.94 vs 11.95 L·min−1·m−2). A high maximal stroke volume is also typical of adult endurance athletes, but reported values in the range of 85–100 mL·m−2 are greater than found in this study of child cyclists (76 mL·m−2) (3,4,12,39). Maximal stroke volume in adult endurance athletes is typically about 40–50% higher than that of nonathletic individuals (3,8,12).
It is clear from this and previous studies that ability to generate heart rate at maximal exercise does not contribute to the greater V̇O2max in athletic compared with nonathletic individuals. Most studies of both child and adult endurance athletes have indicated a slightly lower (usually statistically insignificant) maximal heart rate compared with nonathletic subjects. The lower mean maximal heart rate in the child cyclists in this study was considerably less than that of the nonathletes (184 vs 198 bpm in the nonathletes, P < 0.05), an effect of two particularly low values (both 168 bpm) in the cyclists.
No significant difference was observed in maximal rate of peripheral oxygen extraction between the cyclists and nonathletes (mean arteriovenous oxygen difference 13.1 and 13.0 mL·100 mL−1, respectively). Similarly, studies of adult endurance athletes have indicated that average maximal arteriovenous oxygen difference is equal to or only slightly greater than nonathletic subjects (4,8).
As with all cross-sectional studies, it is impossible to determine whether the physiological differences in child athletes and nonathletes in this study reflect an effect of endurance training or genetic preselection. However, it must be assumed that the features of the cyclists must include the influence of training; therefore, it can be inferred from these data that intensive endurance training in prepubertal boys does not improve maximal arteriovenous oxygen difference. This is in accord with the study of 11- to 13-yr-old boys by Eriksson (5), who found no increase in mean maximal arteriovenous oxygen uptake after 16 wk of aerobic training.
The pattern of stroke volume response to progressive exercise may provide insights into the determinants of stroke volume that separate athletes from nonathletes. In both pediatric and adult nonathletic populations, stroke volume during upright exercise rises 30–40% from resting values at low-moderate exercise intensities but then demonstrates a plateau as workload increases (31). In the present study, the cyclists demonstrated an identical pattern, yet the stroke volume curve was displaced upward. A similar response has been described in adult endurance athletes (2,34).
Resting and maximal values of stroke index were both greater in the athletes, suggesting that factors influencing preexercise stroke volume are critical in determining maximal stroke volume (and thus V̇O2max). The importance of factors affecting resting left ventricular filling (plasma volume, intrinsic ventricular size, and heart rate) in establishing differences in resting stroke volume is suggested by the 2-mm mean greater indexed left ventricular end diastolic dimensions in the cyclists in this study. This difference did not achieve statistical significance (P = 0.17), presumably due to the small number of cyclists in this study. Still, as indicated in Figure 2, six of the seven cyclists demonstrated an indexed EDD above the mean of value in the nonathletic group.
Other studies in adults and children have demonstrated that endurance athletes may be characterized by a failure of stroke volume to plateau at moderate-high exercise intensities, with a progressive rise to exhaustion (8,11). A recent study of child runners in this laboratory (tested on a cycle ergometer) demonstrated a similar stroke volume response (26). This pattern of change suggests that factors influencing stroke volume during progressive exercise (skeletal muscle pump function, diastolic filling properties, ventilation bellows, systemic venoconstriction, and response to catecholamines) may be important in determining the high maximal stroke volume that distinguishes athletes from nonathletes. Further studies will be necessary to determine whether sport and testing modality specificity can explain these differences in pattern of stroke volume response.
In summary, this study of a small number of highly trained child cyclists indicates a pattern of hemodynamic response to exercise that is qualitatively similar to that observed in adult endurance athletes. Moreover, the factors that distinguish aerobic fitness of child athletes from nonathletes appear to be comparable to those of their adult counterparts. Child endurance athletes demonstrate a lower V̇O2max (relative to body size) than adult athletes, a reflection of a lesser maximal stroke volume. The findings of this study suggest that maturity-related differences in resting left ventricular diastolic size (i.e., diastolic filling) are responsible for this greater maximal stroke volume in adult endurance athletes. Whether this can be explained by biological factors, training duration and intensity, and/or selection of highly fit athletes with increasing age remains to be explored.
1. Bekaert, I., J. L. Pannier, C. Van De Weghe, J. P. Van Durme, D. L. Clement, and R. Pannier. Non-invasive evaluation of cardiac function in professional cyclists. Br. Heart J. 45:213–218, 1981.
2. Concu, A., and C. Marcello. Stroke volume response to progressive exercise in athletes engaged in different types of training. Eur. J. Appl. Physiol. 66:11–17, 1993.
3. Dibello, V., G. Santoro, L. Talarico, et al. Left ventricular function during exercise in athletes and in sedentary men. Med. Sci. Sports Exerc. 28:190–196, 1996.
4. Ekblom, B., and L. Hermansen. Cardiac output in athletes. J. Appl. Physiol. 25:619–625, 1968.
5. Erikkson, B. O. Physical training, oxygen supply, and muscle metabolism in 11–13 year old boys. Acta Physiol. Scand. Suppl. 384:1–48, 1972.
6. Fagard, R., A. Aubert, R. Lysens, J. Staessen, L. Van Hees, and A. Amery. Noninvasive assessment of seasonal variations in cardiac structure and function in cyclists. Circulation 67:896–901, 1983.
7. Faria, I. E., E. W. Faria, S. Roberts, and D. Yoshimura. Comparison of physical and physiological characteristics in elite young and mature cyclists. Res. Q. Exerc. Sport 60:388–395, 1989.
8. 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.
9. Gutgesell, H. P., and C. M. Rembold. Growth of the human heart relative to body surface area. Am. J. Cardiol. 65:662–668, 1990.
10. Gutin, B., N. Mayers, J. A. Levy, and M. V. Herman. Physiologic and echocardiographic studies of age-group runners. In: Competitive Sports for Children and Youth, E. W. Brown and C. F. Branta (Eds.). Champaign, IL: Human Kinetics Publishers, 1988, pp. 117–128.
11. Hanson, J. S., and B. S. Tabakin. Comparison of the circulatory response to upright exercise in 25 normal men and 9 distance runners. Br. Heart J. 27:211–219, 1965.
12. Karpman, V. L. Cardiovascular System and Physical Exercise. Boca Raton, FL: CRC Press, 1987, pp. 140–143.
13. Koch, G. Aerobic power, lung dimensions, ventilatory capacity, and muscle blood flow in 12–16 year old boys with high physical activity. In: Children and Exercise IX, K. Berg and B-O. Eriksson (Eds.). Baltimore: University Park Press, 1980, pp. 99–108.
14. Larsen, S. E., H. S. Hansen, K. Froberg, and J. R. Nielsen. Left ventricular structure and function in young elite cyclists. Pediatr. Exerc. Sci.
, in press.
15. Maron B. J. Structural features of the athlete heart as defined by echocardiography. J. Am. Coll. Cardiol. 7:190–203, 1986.
16. Medved, R., V. Fabecic-Sabadi, and V. Medved. Relationship between echocardiographic values and body dimensions in child swimmers. J. Sports Cardiol. 2:28–31, 1985.
17. Miki, T., Y. Yokota, T. Seo, and M. Yokoyama. Echocardiographic findings in 104 professional cyclists with follow-up study. Am. Heart J. 127:898–905, 1994.
18. Nishimura, T., Y. Yamada, and C. Kawai. Echocardiographic evaluation of long-term effects of exercise on left ventricular hypertrophy and function in professional bicyclists. Circulation 61:832–840, 1980.
19. Obert, P., F. Stecken, D. Coureix, A-M. Lecoq, and P. Guenon. Effect of long-term intensive endurance training on left ventricular structure and diastolic function in prepubertal children. Int. J. Sports Med. 19:149–154, 1998.
20. Ozer, S., E. Cil, G. Baltaci, N. Ergun, and S. Ozme. Left ventricular structure and function by echocardiography in childhood swimmers. Jpn. Heart J. 35:295–300, 1994.
21. Rowland, T. W. Developmental Exercise Physiology. Champaign, IL: Human Kinetics Publishers, 1996, pp. 97–116.
22. Rowland, T. W. Does peak VO2
in children? Med. Sci. Sports Exerc. 25:689–693, 1993.
23. Rowland, T. W. Trainability of the cardiorespiratory system during childhood. Can. J. Sports Sci. 17:259–263, 1992.
24. Rowland, T. W. The “trigger hypothesis” for aerobic trainability: a 14-year follow-up. Pediatr. Exerc. Sci. 4:1–9, 1997.
25. Rowland, T. W., B. C. Delaney, and S. F. Siconolfi. “Athlete’s heart” in prepubertal children. Pediatrics 79:800–804, 1987.
26. Rowland, T., D. Goff, B. Popowski, P. Deluca, and L. Ferrone. Cardiac responses to exercise in child distance runners. Int. J. Sports Med. 19:385–390, 1998.
27. Rowland, T., G. Kline, D. Goff, L. Martel, and L. Ferrone. One-mile run performance and cardiovascular fitness in children(Abstract). Med. Sci. Sports Exerc. 30:(Suppl.)S304, 1998.
28. Rowland, T., G. Kline, D. Goff, L. Martel, and L. Ferrone. Physiological determinants of maximal aerobic power in healthy 12-year old boys. Pediatr. Exerc. Sci. 11:317–326, 1999.
29. Rowland, T., E. Melanson, B. Popowski, and L. Ferrone. Test-retest reproducibility of maximal cardiac output by Doppler echocardiography. Am. J. Cardiol. 81:1228–1230, 1998.
30. Rowland, T., and B. Popowski. Comparison of bioimpedance and Doppler cardiac output during exercise in children (Abstract). Pediatr. Exerc. Sci. 9:188, 1997.
31. Rowland, T., B. Popowski, and L. Ferrone. Cardiac responses to maximal upright cycle exercise in healthy boys and men. Med. Sci. Sports Exerc. 19:1146–1151, 1997.
32. Rowland, T. W., V. B. Unnithan, N. G. Macfarlane, N. G. Gibson, and J. Y. Paton. Clinical manifestations of the “athlete’s heart” in prepubescent male runners. Int. J. Sports Med. 8:515–519, 1994.
33. Schairer, J. R., D. Briggs, and T. Kono. Left ventricular function immediately after exercise in elite cyclists. Cardiology 79:284–289, 1991.
34. Schairer, J. R., P. D. Stein, S. Keteyian, et al. Left ventricular response to submaximal exercise in endurance-trained athletes and sedentary adults. Am. J. Cardiol. 70:930–933, 1992.
35. Shepherd, T. A., P. A. Eisenman, H. D. Ruttenberg, T. D. Adams, and S. C. Johnson. Cardiac dimensions of highly trained prepubescent boys (Abstract). Med. Sci. Sports Exerc. 20:(Suppl.)53, 1988.
36. Skaerpe, T., L. Hegrenaes, and H. Ihlen. Cardiac output. In: Doppler Ultrasound in Cardiology, L. Hatle, and B. Angelsen (Eds.). Philadelphia: Lea & Febiger, 1982, pp. 306–320.
37. Slaughter, M. H., T. G. Lohman, R. A. Boileau, et al. Skinfold equations for estimation of body fatness in children and youth. Hum. Biol. 60:709–723, 1988.
38. Telford, R. D., I. G. McDonald, L. B. Ellis, M. H. D. Chennells, E. R. Sandstrom, and P. J. Fuller. Echocardiographic dimensions in trained and untrained 12-year old boys and girls. J. Sports Sci. 6:49–57, 1988.
39. Tomai, F., M. Ciavolella, A. Gaspardone, et al. Peak exercise left ventricular performance in normal subjects and in athletes assessed by first-pass radionuclide angiography. Am. J. Cardiol. 70:531–535, 1992.
40. Wolfe L. A., D. A. Cunningham, and D. R. Boughner. Physical conditioning effects on cardiac dimensions: A review of echocardiographic studies. Can. J. Appl. Sport Sci. 11:66–79, 1986.