Cardiac dynamics during upright cycle exercise in boys

Cardiac Dynamics During Upright Cycle Exercise in BoysTHOMAS ROWLAND1* AND JANET WHATLEY BLUM2

1Department of Pediatrics, Baystate Medical Center,Springfield, Massachusetts2Department of Movement Science, Sport, and Leisure Studies, WestfieldState College, Westfield, Massachusetts

ABSTRACT Insight into the cardiac responses to exercise necessitates anunderstanding of both physiological data and cardiac dimensional changes.This study was designed to assess cardiovascular alterations during progres-sive upright cycle exercise in ten healthy 10–12-year-old boys. Doppler echo-cardiography was used to estimate stroke volume, and 2-dimensional echo-cardiography was used to evaluate changes in left ventricular systolic anddiastolic dimensions. Test–retest reproducibility was high for both tech-niques. Mean peak stroke index and cardiac index values were 62 ± 12 ml m−2

and 11.79 ± 2.62 L min−1 m−2, respectively. Stroke volume rose by 40% overresting values with early exercise (50 watt work load), but beyond moderateintensities (approximately 50% VO2max) little change was seen. The leftventricular diastolic dimension rose slightly at the onset of exercise and thendeclined slowly. A progressive and more precipitous decline was observed insystolic dimension, resulting in an increase in shortening fraction from 29%to 47%. This was accompanied by a dramatic fall in peripheral vascularresistance from 13.9 to 8.0 units at the onset of exercise. These findingssuggest a close matching of increases in heart rate and systemic venousreturn and imply a significant role of peripheral pump function in the circu-latory responses to exercise in children. The cardiac dynamics observed inthis study mimic those previously described in adult subjects. Am. J. Hum.Biol. 12:749–757, 2000. © 2000 Wiley-Liss, Inc.

The central function of the heart duringexercise is to generate cardiac output,which, according to the traditional concept,serves to deliver oxygen and substrate to ex-ercising muscle. A more comprehensive per-spective, however, views the increase in car-diac output with exercise as an integralcomponent of a two pump system which in-cludes not only the heart but also a periph-eral pump responsible for the return of sys-temic venous blood to the heart. Since it isobligatory that the central and peripheralpump outputs must be equal, any limita-tions on circulation during exercise mightbe defined by constraints on peripheralpump as well as cardiac function. Indeed,accumulating evidence suggests that duringexercise, the heart may act to accommodateincreased return from the peripheral pumprather than serving as a primary forwardpump ( Sheriff et al., 1993; Rowell et al.,1996; Laughlin and Schrage, 1999).

Understanding these mechanics requiresnot only an appreciation of hemodynamicdata but also an understanding of changesin cardiac ventricular size during exercise.Specifically, the roles of ventricular preload,contractility, and afterload can only be clari-fied if accurate measures of left ventriculardiastolic and systolic size can be obtained.

Studies in adults utilizing radionuclideangiography and M-mode echocardiographyshow that during a progressive cycle test inthe upright position, the left ventricular enddiastolic (LVED) dimension initially demon-strates a small rise, followed by a slight butprogressive decline (Jeul et al., 1981; Craw-ford et al., 1985; Higginbotham et al., 1986;

*Correspondence to: Thomas Rowland, Department of Pediat-rics, Baystate Medical Center, Springfield, MA 01199.

Received 10 November 1999; Revision received 4 January2000; Accepted 9 January 2000

AMERICAN JOURNAL OF HUMAN BIOLOGY 12:749–757 (2000)

© 2000 Wiley-Liss, Inc.

PROD #M99088R

Younis et al., 1990). These studies have in-dicated that the left ventricular end systolic(LVES) dimension falls more precipitously,resulting in an increase in ejection fractionor shortening fraction. These dimensionalchanges are accompanied by an early 40%rise in stroke volume and then plateau toexhaustion. These findings suggest that be-yond the low-intensity stages of exercise,the heart rate increases to match the aug-mented systemic venous return and main-tain a relatively stable left ventricular dia-stolic size. The increase in myocardialcontractility may then serve to maintainstroke volume in the face of a decrease inventricular ejection time as the heart raterises. This interpretation is consistent withthe concept that it is the peripheral pumpwhich may “drive” the circulation during ex-ercise and suggests the possibility that athigh exercise intensities circulatory flowcould be limited by peripheral as well as car-diac factors.

In supine or semisupine exercise, theseresponses are often different (Poliner et al.,1980). The initial rise in stroke volume maybe absent or attenuated, and LVED size hasbeen observed to be stable or increase withexercise. Increases in left ventricular ejec-tion fraction are less in the supine comparedto the upright position. These characteris-tics presumably reflect the absence of gravi-tational influences on systemic venous re-turn.

Whether these responses are also seen inpediatric subjects is unclear. No previousstudy has measured alterations in left ven-tricular dimensions during upright exercisein children. Using M-mode echocardiogra-phy, Rowland et al. (2000) demonstrated agradual 5% decrease in end diastolic ven-tricular dimension in 7–14-year-old chil-dren during exhaustive semisupine cycling.A greater fall in systolic dimension resultedin an increase in shortening fraction from amean of 37% to 53%. Three studies in chil-dren during supine cycling also indicated aprogressive decline in diastolic size, a moreexaggerated fall in systolic size, and an in-crease in shortening fraction or ejectionfraction (DeSouza et al., 1984; Parrish et al.,1985; Oyen et al., 1987).

The utility of echocardiography duringupright exercise is limited by motion arti-fact. In this study, this difficulty was mini-mized by measuring diastolic and systolicdimensions on video tapes of two-dimen-

sional echocardiograms, using selectedframes that satisfied specific measurementcriteria. Using this information, we soughtto (1) describe the normal ventricular di-mensional changes with progressive exer-cise during upright cycling in children, (2)characterize the cardiac responses to exer-cise in this age group by relating dimen-sional changes to hemodynamic data, and(3) compare these findings with those previ-ously reported in adults as well as duringsemisupine and supine exercise in children.In addition, information was sought regard-ing the reliability of the echocardiographictechniques utilized in this study.

METHODS

Ten healthy boys aged 10–12 years old(mean age 11.8 ± 0.8 years) who were chil-dren of medical center employees were re-cruited for exercise testing. None was in-volved in regular athletic training, but 7had participated on community sportsteams over the past three months (mainlybasketball and soccer). Parents were askedto estimate their child’s level of physical ac-tivity on a scale of 1 (inactive: watches TV,reads or does homework after school) to 5(very active: involved in sports, dislikes sed-entary activities). The mean value for thegroup was 3.9 (±0.8). By questionnaire,none of the subjects had appearance of fa-cial or pubic hair and/or voice change, sug-gesting prepubertal status.

Informed consent was obtained from theparents of all children. This study was re-viewed and approved by the InstitutionalReview Board of the Baystate Medical Cen-ter.

Subjects were each tested on two separateoccasions 1–2 weeks apart by identical ex-ercise protocols. Prior to each test, heightwas measured with a stadiometer andweight, with a calibrated balance-beamscale. Body surface area was estimated froma nomogram derived from the formula ofDubois and Dubois (1916). To characterizethe subject population, the triceps and sub-scapular skinfold measurements were mea-sured using Holtain (Croswell, England)calipers by the same examiner. Averages ofthree measurements were summed to cre-ate a skinfold score.

Progressive upright exercise to exhaus-tion was performed on a mechanicallybraked Monark ergometer in an ambienttemperature of 20–21°C with moderate hu-

750 T. ROWLAND AND J.W. BLUM

midity (40–60%). The seat height was ad-justed to create a small knee angle (approxi-mately 15°) at full extension. Initial andincremental loads were 25 watts, with aconstant 50 rpm cadence. Stage durationwas 3 min.

Heart rate was measured by electrocar-diogram. Gas exchange variables were de-termined with standard open circuit tech-niques using a Q-Plex Cardio-PulmonaryExercise System (Quinton Instrument Com-pany, Seattle, WA). Subjects breathedthrough a 94-ml dead space Rudolph valve,with expired samples drawn from a 6-L mix-ing chamber. Oxygen and carbon dioxidecontents of expired samples were analyzedby zirconia oxide and infrared analyzers, re-spectively. A pneumotachometer in the ex-piratory line measured minute ventilation.The system was calibrated before each testwith standard gases of known oxygen andcarbon dioxide concentration.

Average values for oxygen uptake (VO2),carbon dioxide production (VCO2), VCO2/VO2 (respiratory exchange ratio, RER), andminute ventilation (VE) were calculatedover 15-sec intervals. Peak VO2 was identi-fied as the mean of the two highest values inthe final minute of exercise, and it was alsoconsidered equivalent to VO2max if the sub-ject exhibited signs of exhaustion with amaximal heart rate over 185 bpm and maxi-mal RER exceeding 1.00.

Stroke volume was estimated using stan-dard Doppler echocardiographic techniques(Calafiore and Stewart, 1990). Ascendingaortic blood velocity was measured with a1.9 MHz continuous wave transducer di-rected caudad from the suprasternal notch.The integral of velocity over time (velocity–time integral, VTI) for each beat was as-sessed by tracing the velocity curve contourboth on-line and off-line. VTI values at rest,during the last minute of each stage, and inthe final 30 sec of exercise were determinedby averaging the 3–10 curves with the high-est values and most distinct spectral enve-lopes.

Stroke volume was estimated by multi-plying the mean VTI by the maximal cross-sectional area of the ascending aorta at rest.The latter was calculated from the mean of5–10 measurements of the aortic diameterat the level of sino-tubular junction (inneredge to inner edge) measured with two-dimensional echocardiography (parasternallong axis view) with the subject in the sit-

ting position. Cardiac output was estimatedas the product of heart rate and stroke vol-ume, and both cardiac output and strokevolume were related to body surface area(cardiac index, stroke index). Arteriovenousoxygen difference was calculated by divid-ing oxygen uptake by cardiac output.

Echocardiographic determination of leftventricular chamber size during exercise ishampered by the variability of heart posi-tion and respiratory interference with ultra-sound views. In this study a new techniqueof echocardiographic imaging was developedwhich allowed measurement of systolic anddiastolic ventricular dimensions throughoutprogressive exercise. These dimensionswere estimated directly by two-dimensionalechocardiography (parasternal long axisview) using a 3.5 MHz transducer (HewlettPackard Sonos 1000, Andover, MA). Mea-surements were made at rest, at the begin-ning of the third minute of each work stage,and 1 min before termination of exercise.Echocardiograms were videotaped, andmeasurements of diastolic and systolic sizewere made off-line. The following conven-tion was introduced for ventricular diam-eters with this technique: all measurementswere made perpendicular to the left ven-tricular long axis, only in frames in whichthe left ventricular cavity, both aortic andmitral valve leaflets, and the aortic rootwere visible. Measurements defined as dia-stolic were taken in the first frame of mitralvalve closure at the onset of ventricular con-traction, and ventricular systolic diameterwas identified in the last frame before mi-tral valve opening at the termination of ven-tricular contraction. Measurements weretaken as the shortest distance from septal tofree wall endocardial surface, 1–2 cm distalto the closed mitral valve, perpendicular tothe left ventricular long axis. Mean valueswere calculated from the 5–10 measure-ments obtained at each stage.

Auscultatory blood pressure was mea-sured in the left arm at rest and in the finalminute of each submaximal work stage,only during the second test. Mean arterialpressure was determined as one-third of thepulse pressure plus the diastolic pressure.Systemic vascular resistance was calculatedas the quotient of mean arterial pressuredivided by cardiac output.

Reproducibility of stroke volume mea-surements and ventricular dimensions inthe two tests was assessed by repeated mea-

CARDIAC DYNAMICS DURING EXERCISE IN BOYS 751

sures ANOVA (trial × workload), means andstandard deviations of individual differ-ences, and coefficient of variation (standarddeviation of the differences divided by themean of the two values). Statistical signifi-cance was defined as P < 0.05.

RESULTS

Descriptive characteristics (mean ± stan-dard deviation) of the subjects on the initialtest were as follows: weight 44.8 ± 8.1 kg,height 149 ± 7 cm, body surface area 1.37 ±0.16 m2, and skinfold score 25.4 ± 11.7 mm.Averages for peak VO2 on the two tests were46.0 ± 7.1 and 44.9 ± 7.7 ml kg−1 min−1 (P >0.05). Comparable mean values for peakheart rate (190 ± 10 and 188 ± 7 bpm), peakRER (0.98 ± 0.04 and 0.96 ± 0.04), and en-durance time (14.20 ± 2.21 and 13.51 ± 1.94min) indicated similar efforts at peak exer-cise in both tests (P > 0.05 for all three com-parisons).

Because of the time required for both 2-di-mensional and Doppler ultrasound mea-surements, the point of maximal, exhaus-tive exercise had to be anticipated 1 min inadvance. Inability to accurately estimatethis time by the examiner in all cases re-sulted in a failure of subjects to consistentlyachieve a true exhaustive effort. Of the 20tests, 11 failed to reach criteria for VO2max.However, by subjective observation, a high-intensity, near-maximal effort was consid-ered to have been provided on all tests.

Repeated measures ANOVA indicated nomain effect of trial nor of trial–workload in-teraction. Mean values for hemodynamicand echocardiographic variables during thetwo tests at pre-exercise, 50 watt submaxi-mal workload, and peak exercise are out-lined in Table 1. These average measureswere highly reliable. No significant groupdifferences were observed in the two ses-sions for submaximal or peak heart rate,stroke volume, cardiac output, arteriove-nous oxygen difference, ventricular systolicand diastolic dimensions, or ventricularshortening fraction. Peak stroke index was62 ± 12 and 63 ± 12 ml m−2 on tests 1 and 2,respectively, and peak cardiac index valueswere 11.79 ± 2.62 and 11.88 ± 2.53 L min−1

m−2 (P > 0.05).Average differences between tests, the

standard deviation of these differences, andthe coefficient of variation for stroke volumeand echocardiographic variables are de-scribed in Table 2. Individual differences forpeak stroke volume for the two tests ranged

from −7 to +8 ml, with a single outlyingvalue of +18 ml. When this latter value wasincluded, the coefficient of variation was8.1%. The coefficient was 5.7% when theoutlying determination was ignored. Amongthe echocardiographic measures, paired val-ues for left ventricular diastolic diameter forindividuals proved most reliable, with coef-ficients of variation of 3.2% and 6.0% at sub-maximal and peak exercise, respectively.

Stroke volume demonstrated an initialrise until approximately 50% peak VO2, fol-lowing which little change was observed topeak exercise (Fig. 1). The differences be-tween maximal and pre-exercise valueswere statistically significant in both trials.The ratios of mean peak to pre-exercise val-ues were 1.43 and 1.48 for the two tests.

TABLE 1. Mean (standard deviation) values for testand retest*

Test 1 Test 2Endurance time (min) 14.20 (2.33) 13.51 (2.05)Peak VO2 (ml kg−1 min−1) 46.0 (7.5) 44.9 (8.1)Peak RER 0.98 (0.04) 0.96 (0.04)

Hemodynamic

Heart rate (bpm)Pre-exercise 83 (12) 83 (16)50 watts 125 (10) 129 (8)Peak 190 (11) 188 (7)

Stroke volume (ml)Pre-exercise 58 (9) 58 (12)50 watts 81 (13) 80 (11)Peak 83 (15) 86 (15)

Cardiac output (L min−1)Pre-exercise 4.77 (0.95) 4.87 (1.34)50 watts 10.08 (1.86) 10.24 (1.44)Peak 15.82 (3.21) 16.09 (3.06)

Arteriovenous oxygendifference (ml 100ml−1)

Pre-exercise 7.7 (2.1) 6.4 (1.5)50 watts 10.3 (2.0) 10.4 (2.1)Peak 13.1 (2.2) 12.5 (2.2)

Echocardiographic

Left ventricular diastolicdimension (cm)


Left ventricular systolicdimension (cm)


Left ventricular shorteningfraction (%)


*All differences in these comparisons fail to reach statisticalsignificance (P > 0.05).


The pattern of response of the left ven-tricular diastolic dimension to increasingwork was consistent in the two tests (Fig. 2).In both trials, a small initial rise was ob-served at onset of exercise, followed by agradual decline. On both tests the resultingvalue at peak exercise was significantly lessthan observed at pre-exercise (3.75 vs 3.99cm; 3.83 vs 4.01 cm; P < 0.05).

The fall in systolic dimension, on theother hand, was more precipitous and con-sistent throughout each test (Fig. 2). Conse-quently, left ventricular shortening fractionrose as work increased (Fig. 3). The overallrise was from 29.4% to 46.6% in test 1 andfrom 31.0% to 47.4% in test 2. The rate ofrise was most exaggerated at the onset ofexercise in both trials, a reflection of boththe initially increased pre-load filling

(greater diastolic dimension) and fall in sys-tolic dimension.

Mean systemic vascular resistance fellprecipitously at onset of exercise, from 13.9± 4.4 units at pre-exercise to 8.0 ± 1.5 unitsat 25 watt work load. Only small decreaseswere subsequently observed (6.8 ± 1.1 and6.0 ± 1.2 units at 50 and 75 watts, respec-tively).

DISCUSSION

The level of cardiovascular fitness in thisgroup of subjects is consistent with that de-

TABLE 2. Mean (standard deviation) differencesbetween tests 1 and 2 and coefficient of variation (CV)

Mean difference CVStroke volume (ml)

Pre-exercise +1 (11) 19.1%50 watts −1 (8) 9.7Peak +3 (7) 8.1

LV diastolic dimension (cm)Pre-exercise −0.02 (0.26) 6.550 watts −0.01 (0.13) 3.2Peak +0.08 (0.23) 6.0

LV systolic dimension (cm)Pre-exercise −0.05 (0.33) 11.850 watts −0.08 (0.25) 10.8Peak +0.02 (0.28) 13.8

LV shortening fraction (%)Pre-exercise +1.6 (6.3) 20.850 watts +2.2 (5.5) 13.1Peak +0.8 (6.7) 14.4

Fig. 1. Stroke volume response to progressive exer-cise during test 1 (solid line) and test 2 (dashed line).

Fig. 2. Left ventricular diastolic and systolic dimen-sional changes during test 1 (solid line) and test 2(dashed line).

Fig. 3. Left ventricular shortening fraction changesduring test 1 (solid line) and test 2 (dashed line).


scribed in other studies involving healthychildren. The peak VO2 of 46.0 ml kg−1

min−1 is typical of that described in 12-year-old boys during cycle testing (Rowland,1996). Reports of maximal cardiac index inprepubertal boys have ranged from 9.2 to12.0 L min−1 m−2 and maximal stroke indexfrom 46 to 64 ml m−2 (see Rowland, in press,for review), consistent with the overall av-erage values of 11.84 L min−1 m−2 and 63 mlm−2, respectively, in the present study.These observations support the conclusionthat the subjects in this study achievednear-maximal exercise levels, although cri-teria for VO2max were not consistently met.They also imply that the pattern of hemo-dynamic responses to exercise in this studyshould therefore be expected to be typicalfor healthy, non-athletically trained 10–12-year-old boys.

A high degree of reliability was observedfor both the Doppler technique for estimat-ing stroke volume and the approach for as-sessing left ventricular chamber size duringexercise with two-dimensional echocardiog-raphy. A group average difference in peakstroke volume of only 1 ml was observed ontest–retest, with a standard deviation of 7ml and a coefficient of variation of 8.1%.When an outlying value was removed fromthe analysis, the standard deviation of thedifferences became 5 ml, with a coefficient ofvariation of 5.7%. These findings are com-parable to a previous investigation of repro-ducibility of maximal stroke volume inyoung adult males in this laboratory (Row-land et al., 1998). In that study a mean dif-ference of 5 ml was observed, with a coeffi-cient of variation of 8.5%. Similar levels ofvariation have been previously described forsubmaximal Doppler (Mouliner et al., 1991)as well as other techniques for estimatingcardiac output during exercise (thermodilu-tion, CO2 rebreathing) (Moore et al., 1992;Esperson et al., 1995).

The two-dimensional echocardiographictechnique introduced in this study was de-signed to provide information regardingventricular size during exercise by avoidingmotion and ventilatory interference. By se-lecting only those video frames that satis-fied specific criteria, it was possible to ob-tain such measurements up to peakexercise. Mean differences for diastolic andsystolic dimensions on retest using this ap-proach were very small. At peak exercise,the individual variability was small to mod-

erate, with coefficients of variation of 6.0%and 13.8% for diastolic and systolic dimen-sions, respectively. The pattern of changefor these measures on the two tests wasidentical. These findings suggest that thisapproach is useful in assessing left ventricu-lar dimensional changes with exercise.

The patterns of change in hemodynamicvariables in this study can be interpretedwithin the construct that (1) the circulationduring exercise comprises a two-pump sys-tem, a central pump (the heart) and a pe-ripheral pump (provided by skeletal musclecontraction, ventilatory bellows, left ven-tricular “suction,” and abdominal musclecontraction), and (2) the function of the twopumps must be equal. Within this model,there are certain observations described inthis study as well as others which must beaccounted for in any attempt to formulate aunified description of the circulatory re-sponses to exercise:

1. Stroke volume rises by 30–50% at theinitiation of upright exercise but then pla-teaus as exercise intensity increases. In thepresent study stroke volume rose initiallyby approximately 40% above pre-exercisevalues, and no significant change was ob-served above an exercise intensity of 50%VO2max. Such a pattern is one of the mostconsistently described phenomena in car-diac exercise physiology. Failure of strokevolume to increase beyond mild to moderateexercise intensities has been described inmultiple studies in children (Eriksson andKoch, 1973; Koch et al., 1997; Rowland etal., 1997; Turley and Wilmore, 1997) as wellas adults (Higginbotham et al., 1986; You-nis et al., 1990).

2. No dramatic changes are observed inleft ventricular diastolic dimension duringprogressive exercise. In this study a small(0.10 cm) rise in diastolic dimension oc-curred at the first workload, followed by aslow gradual decline to peak exercise. At ex-haustion, the average value was signifi-cantly lower than that at rest. This patternof change in diastolic dimension is identicalto that described by others during uprightexercise in adults (Keul et al., 1981; Higgin-botham et al., 1986; Younis et al., 1990).Previous studies in children during supineand semisupine exercise have observed agradual decline in diastolic dimension withincreasing workload but without an earlyrise (Oyen et al., 1987; Rowland et al.,2000). Studies involving supine exercise in


adults have indicated stable (Crawford etal., 1985), increasing (Poliner et al., 1980),and decreasing (Adams et al., 1992) diastol-ic size during progressive exercise.

3. Markers of cardiac contractility (suchas left ventricular shortening fraction) in-crease throughout the course of progressiveexercise, with the greatest rise at the onset.In the present study, average left ventricu-lar shortening fraction increased from ap-proximately 30% to 38% at the first work-load and then continued a slower rise to47% at peak exercise. (It should be recog-nized that these values for shortening frac-tion cannot be related to those traditionallyused in clinical cardiology, as a differentconvention for measurement was utilized.)This increase of 17% is comparable to thatdemonstrated by Crawford et al. (1985)(from 37% to 50%) in adults with uprightexercise. Similar findings have been ob-served in children. Oyen et al. (1987) ob-served a rise in shortening fraction from36% at rest to 47% at maximal exercise dur-ing supine exercise, and Rowland et al.(2000) described an increase from 37% to53% during maximal semisupine cycling.

4. A dramatic decline in peripheral vascu-lar resistance is observed at the onset of ex-ercise. Calculated systemic vascular resis-tance fell from 13.9 units at rest to 8.0 unitat the first work stage in this study, with aslower subsequent decline as work in-creased. Rowland et al. (2000) showed asimilar pattern in children during semisu-pine exercise, with maximal resistance val-ues being less than half those at pre-exercise. Most (68%) of this decline tookplace at the first workload. Equivalent de-clines in systemic vascular resistance withexercise in children have been reported byothers (Eriksson and Koch, 1973; Cyran etal., 1988; Ensing et al., 1988).

In a previous study involving semisupineexercise, we attempted to provide a synthe-sis of the cardiovascular responses to exer-cise in children based on Doppler and two-dimensional echocardiographic measures(Rowland et al., 2000). In the present studythese observations were extended to uprightexercise, with similar conclusions.

The rise in stroke volume at the initiationof upright exercise appears to be related to acombination of increased preload (as indi-cated by the increase in diastolic dimension)and sudden fall in peripheral vascular resis-tance. These changes, in turn, are manifes-

tations of a combination of dramatic in-creases in peripheral vascular conductanceand initiation of skeletal muscle pump func-tion (Sheriff et al., 1993). Such findings sug-gest that the initial responses of the circu-latory system to upright exercise are drivenby the peripheral rather than the centralpump in both adults and children.

As exercise intensity subsequently in-creases, stroke volume remains stable, andonly small changes are observed in left ven-tricular diastolic size and systemic vascularresistance. These observations imply thatthe rise in heart rate closely matches theincreasing volume of systemic venous re-turn, maintaining left ventricular diastolicsize (and thus preload).

Above moderate intensity work, indices ofmyocardial contractility (shortening frac-tion, ejection fraction, peak aortic velocity)demonstrate a continual rise. This patternoccurs at the same time that stroke volumeremains stable. These changes can best beexplained by the importance of catechol-amine-induced myocardial contractility inincreasing the rate of systolic ejection to ac-commodate the progressively shorteningleft ventricular ejection time (Braunwald etal., 1958).

This physiologic “scenario” synthesizedfrom the findings in this and other studiessuggests the primary importance of the pe-ripheral pump in “driving” the circulation ofblood during progressive exercise. The dataindicate that the principal roles of the cen-tral pump are to (1) increase contractionrate to accommodate return flow, and (2)augment contractility to deliver a constantstroke volume as systolic ejection time de-creases. This interpretation bears impor-tance in the search for factors which mightlimit circulatory responses to exercise (andthus delimit exercise capacity), implyingthat such determinants may lie in the pe-ripheral rather than the central componentsof the circulatory system.

These data also provide evidence that thepattern of cardiovascular responses to exer-cise in children mimics that of adults. Mat-urational differences in cardiac functionwith exercise appear to be entirely quanti-tative, related to only growth and body size.Direct comparisons of cardiac responses toexercise in children and adults demon-strated no significant differences in size-relative cardiac output, stroke volume, orpattern of change of these variables (Row-


land et al., 1997). Similarly, the patterns ofventricular dimensional changes duringprogressive upright exercise observed in theboys in the present study are not differentthan those previously described in adults.

The findings in this study support the useof Doppler and two-dimensional echocardio-graphic techniques for the safe and accurateestimation of cardiac variables during maxi-mal exercise testing in children. Potentialdrawbacks to the Doppler method includenonuniformity of laminar ascending aorticflow, change in aortic root diameter with ex-ercise, and angular variability of the supra-sternal probe and ascending aorta (Cala-fiore and Stewart, 1990). However, thereliability of this method and favorable com-parisons of cardiac values with those ob-tained by other methods suggest that thesesources of error are minimal. Similarly, thereproducibility of left ventricular dimen-sions during exercise using the approach foruse of two dimensional echocardiography inthis study suggests that motion artifact maybe effectively minimized using this tech-nique.

ACKNOWLEDGMENT

The authors are indebted to Dr. BabettePluim for her helpful review of this manu-script.

LITERATURE CITED

Adams KF, McAllister SM, El-Ashmawy H, Atkinson S,Koch G, Sheps DS. 1992. Interrelationships betweenleft ventricular volume and output during exercise inhealthy subjects. J Appl Physiol 73:2097–2104.

Braunwald E, Sarnoff SJ, Stainsby WN. 1958. Determi-nants of duration and mean rate of ventricular ejec-tion. Circ Res 6:319–325.

Calafiore P, Stewart WJ. 1990. Doppler echocardio-graphic quantitation of volumetric flow rate. CardiolClin 8:191–202.

Crawford MH, Petru MA, Rabinowitz C. 1985. Effect ofisotonic exercise on left ventricular volume during ex-ercise. Circulation 72:1237–1243.

Cyran SE, James FW, Daniels S, Mays W, Shukla R,Kaplan S. 1988. Comparison of the cardiac output andstroke volume response to upright exercise in childrenwith valvular and subvalvular aortic stenosis. J AmColl Cardiol 11:651–658.

DeSouza M, Scaffer MS, Gilday DL, Rose V. 1984. Ex-ercise radionuclide angiography in hyperlipidemicchildren with apparently normal hearts. Nucl MedCommun 5:13–17.

DuBois D, DuBois EF. 1916. Clinical calorimetry, tenthpaper. A formula to estimate approximate surfacearea if height and weight be known. Arch Intern Med17:863–866.

Ensing GJ, Heise CT, Driscoll DJ. 1988. Cardiovascularresponse to exercise after the Mustard operation for

simple and complex transposition of the great vessels.Am J Cardiol 62:617–622.

Eriksson BO, Koch G. 1973. Effect of physical trainingon hemodynamic response during submaximal andmaximal exercise in 11–13-year-old boys. Acta Phys-iol Scand 87:27–39.

Esperson K, Jensen EW, Rosenberg D, Thomsen JK,Eliasen K, Olsen NV, Kanstrup IL. 1995. Comparisonof cardiac output measurement techniques: thermodi-lution, Doppler, CO2 rebreathing, and direct Fickmethod. Acta Anaesthesiol Scand 39:245–251.

Higginbotham M, Morris KG, Williams RS, McHale PA,Coleman RE, Cobb FR. 1986. Regulation of stroke vol-ume during submaximal and maximal upright exer-cise in normal man. Circ Res 58:281–291.

Keul J, Dickhuth HH, Simon G, Lehmann M. 1981. Ef-fect of static and dynamic exercise on heart volume,contractility, and left ventricular dimensions. CircRes 48(Suppl 1):162–170.

Koch G, Mobert J, Oyen E-M. 1997. Cardiovascular ad-justment to supine versus seated posture in prepuber-tal boys. In: Armstrong N, Kirby B, Welsman J, edi-tors. Children and exercise XIX. London: E&FN Spon.p 424–428.

Laughlin MH, Schrage WG. 1999. Effects of muscle con-traction on skeletal muscle blood flow: when is there amuscle pump? Med Sci Sports Exerc 31:1027–1035.

Moore R, Sansores R, Guimond V, Abboud R. 1992.Evaluation of cardiac output by thoracic electrical bio-impedance during exercise in normal subjects. Chest102:448–445.

Mouliner L, Venet I, Schiller NB, Kurtz TW, Morris RC,Sebastian A. 1991. Measurement of aortic blood flowby Doppler echocardiography: day to day variabilityin normal subjects and applicability in clinical re-search. J Am Coll Cardiol 17:1326–1333.

Oyen EM, Ignatzy K, Ingerfeld G, Brode P. 1987. Echo-cardiographic evaluation of left ventricular reserve innormal children during supine bicycle exercise. Int JCardiol 14:145–154.

Parrish MD, Boucek RJ, Burger J, Artman MF, PartainCL, Graham TP. 1985. Exercise radionuclide ven-triculography in children: normal values for exercisevariables and right and left ventricular function. BrHeart J 54:509–516.

Poliner LR, Dehmer GJ, Lewis SE, Parkey RW, Blom-quist CG, Willerson JT. 1980. Left ventricular perfor-mance in normal subjects: a comparison of the re-sponses to exercise in the upright and supinepositions. Circulation 62:528–534.

Rowell LB, O’Leary DS, Kellogg DL. 1996. Integrationof cardiovascular control systems in dynamic exercise.In: Handbook of physiology, Section 12, Integration ofmotor, circulatory, respiratory, and metabolic controlduring exercise. Bethesda, MD: American Physiologi-cal Society. p 770–838.

Rowland TW. 1996. Developmental exercise physiology.Champaign, IL: Human Kinetics Publishers. p 73–96.

Rowland TW. Cardiovascular function. In: ArmstrongN, Van Michelen W, editors. Paediatric exercise sci-ence and medicine. Oxford: Oxford University Press(in press).

Rowland TW, Melanson EL, Popowski BE, Ferrone LC.1998. Test–retest reproducibility of maximum cardiacoutput by Doppler echocardiography. Am J Cardiol1228–1229.

Rowland TW, Popowski B, Ferrone L. 1997. Cardiac re-sponses to maximal upright cycle exercise in healthyboys and men. Med Sci Sports Exerc 29:1146–1151.


Rowland T, Potts J, Potts T, Sandor G, Goff D, FerroneL. 2000. Cardiac responses to progressive exercise innormal children: a synthesis. Med Sci Sports Exerc31:253–259.

Sheriff DD, Rowell LB, Scher AM. 1993. Is rapid rise invascular conductance at onset of dynamic exercisedue to muscle pump? Am J Physiol 265:H1227–H1234.

Turley KR, Wilmore JH. 1997. Cardiovascular re-sponses to submaximal exercise in 7- to 9-year-old boys and girls. Med Sci Sports Exerc 29:824–832.

Younis LT, Melin JA, Robert AR, Detry JMR. 1990. In-fluence of age and sex on left ventricular volumes andejection fraction during upright exercise in normalsubjects. Eur Heart J 11:916–924.


Cardiac dynamics during upright cycle exercise in boys

Documents

Transcript of Cardiac dynamics during upright cycle exercise in boys