- Email: [email protected]
Effect of long-term high intensity aerobic training on left ventricular volume during maximal upright exercise
JACC Vol . 14, No . August 1989 :364-71
364
Effect of Long-Term High Intensity Aerobic Training on Left Ventricular Volume During Maximal Upright Exercise LEONARD E . GINZTON, MD, FACC, RICHARD CONANT, PHD, MARIANNE BRIZENDINE, RN, MICHAEL M . LAKS, MD, FACC Torrance, California
The purpose of this study was to determine whether high intensity, long-term aerobic training causes the left ventricle to develop different mechanisms for increasing cardiac output during submaximal and maximal upright bicycle exercise. Fifteen competitive collegiate long distance runners and 14 healthy sedentary adults were studied with use of subcostal view four chamber two-dimensional echocardiography at rest and during and at peak maximal upright bicycle exercise . At rest, the athletes had a larger end-diastolic volume index (85 ± 14 ml/m ) ( mean ± 1 SD) than that of the sedentary adults (6 ± 14 MI/M ) and a larger end-systolic volume index (37 ± 11 versus 1 ± 6 MI/M ) . During low and moderate intensity exercise, end-diastolic and stroke volume indexes increased in both groups, but at high intensity exercise and at peak exercise the end-diastolic volume index of both groups decreased significantly below
Athletes engaged in long-term high intensity aerobic training are known to have a dilated left ventricle at rest (1-4) . However, the mechanisms of the left ventricular dilation, and the possibility that the mechanisms for increasing cardiac output during exercise differ in sedentary persons and athletes, have long been a subject of dispute (4-8) . Elucidation of these mechanisms would have implications both for exercise training and for understanding the role of exercise in myocardial hypertrophy and adaptation to left ventricular volume overload states . Some investigators (7,9) have reported that an increased cardiac output during exercise is achieved by increasing left ventricular end-diastolic volume
From the Division of Cardiology, Department of Medicine, HarborUniversity of California at Los Angeles Medical Center and the Department of Medicine, University of California at Los Angeles School of Medicine, Torrance, California . Manuscript received August 5, 1988 : revised manuscript received February 15, 1989, accepted March 3, 1989 . Address for reprints : Leonard E . Ginzton, MD, Division of Cardiology, Harbor-University of California at Los Angeles Medical Center, 1000 West Carson Street, Torrance, California 90509 . (0 1989 by the American College of Cardiology
rest value (athletes, 61 ± 14 ; sedentary subjects, 46 ± 10 mI/mz , both p < 0 .001 compared with rest) . Reflecting the decreased end-diastolic volume index, at peak exercise, the stroke volume index had decreased from intermediate exercise values in both groups and was not different from rest values. Therefore, although long distance runners have a dilated left ventricle at rest, they utilize the same mechanisms as sedentary adults for increasing cardiac output during upright dynamic exercise . At low and moderate level exercise, the Frank-Starling mechanism is a dominant mechanism for increasing cardiac output, but at peak exercise, probably because of reduced diastolic left ventricular filling, enhanced contractility is the major mechanism for maintaining stroke volume . (J Am Coll Cardiol 1989 ;14:364-71)
and stroke volume, especially in highly trained endurance athletes (4) . These data have been cited as evidence that the Frank-Starling mechanism is utilized preferentially by athletes during exercise and that athletes and sedentary subjects, therefore, have fundamentally different mechanisms for increasing cardiac output during exercise . However, other investigators (5,10) have reported that stroke volume increases early in exercise in highly trained athletes but does not continue to increase with further increases in exercise load to maximal exercise . These investigators have concluded that increased contractility, rather than the FrankStarling mechanism, is the predominant mechanism by which cardiac output is increased during exercise in athletes . They have proposed that the left ventricular dilation observed in athletes may be secondary to a bradycardia at rest that causes the left ventricle to dilate to maintain a normal cardiac output at rest (10-13) . In many previous studies of cardiovascular dynamics during exercise in athletes, exercise testing has often been carried out in the supine position (9,14,15), which poses 0735-1097/89/$3 .50
JACC Vol . 14, No . August 1989 :364-71
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
different preload and afterload conditions on the heart from those posed by upright exercise . Also, left ventricular volume estimates have often been made by techniques that require up to several minutes of data gathering, which may obscure the effects of rapidly changing left ventricular loading conditions and function, especially at maximal exercise (15,16) . Therefore, it has been difficult to apply these data meaningfully to maximal exercise in the upright position . Our laboratory has reported (17) that subcostal view two-dimensional echocardiography is a useful, accurate method for obtaining left ventricular volume measurements on a beat by beat basis during and at the peak of maximal upright bicycle exercise . Therefore, this technique seemed applicable to the study of changes in left ventricular volume in endurance-trained athletes . We have observed that in sedentary, healthy adults, the end-diastolic volume decreases at the peak of symptom-limited upright bicycle exercise and that, at peak exercise, stroke volume is unchanged from control . We hypothesized that endurance athletes would have a similar response during exercise . If the athletes' end-diastolic volume decreases at peak exercise, this would suggest that their increased exercise cardiac output is primarily related to increased heart rate and myocardial contractility, and that they do not utilize the Frank-Starling mechanism to adapt to maximal upright exercise . Therefore, the purpose of this study was to compare changes in left ventricular volume during maximal upright bicycle exercise in competitive long distance runners and sedentary normal adults .
Methods Study groups . Two groups of subjects were studied . The athletic group was made up of 15 competitive long distance runners, all of whom were members of their collegiate cross-country running team . Their age was 19 ± 1 years (mean ± I SD) ; 9 were male and 6 female . The runners had averaged training distances of 116 ± 5 km/week for an average of 4 .4 ± 1 years . All had exceeded 70 km/week in training for at least the previous 8 consecutive weeks . The control group comprised 14 sedentary, healthy adults whose mean age was 8 ± 6 years ; 10 were male and 4 female . None of the control group had participated in any competitive or regular conditioning activity for >5 years . None of the subjects in either group had any clinical or historical evidence of cardiovascular disease or hypertension . None was taking any medication and all had a normal cardiac physical examination and a normal 1 lead electrocardiogram (ECG) both at rest and at the peak of maximal upright bicycle exercise . All subjects were studied according to a research protocol approved by the Research Committee and Human Subjects Committee of Harbor-University of California at Los Ange-
365
les Medical Center and all subjects gave written informed consent before participating in the study . Exercise protocol . All subjects were studied at the same time of day (mid afternoon) . They were exercised in the upright position on a bicycle ergometer with continuous ECG monitoring . A 1 lead ECG and sphygmomanometric blood pressure were obtained at rest, every 3 min during exercise and at peak exercise . Exercise was begun at a zero work load and increased by 75 kilopond meters (kp-m)/min each minute until the subject adamantly refused further exercise because of exhaustion . Echocardiographic protocol . Subcostal four chamber view two-dimensional echocardiograms were obtained with use of a wide angle mechanical scanner (Advanced Technologies Laboratories Mark III) . The subcostal four chamber view was recorded on videotape with the subject sitting upright at rest, every 3 min during exercise and from approximately 15 s before the attainment of peak exercise until peak exercise was reached . No post-exercise images were used for measurement of left ventricular volumes . To obtain reproducible views of the left ventricle, the following, previously reported protocol (17) was employed . The handheld transducer was placed in the subxiphoid position just below or to the left of the xiphoid process and gently pushed inward and upward with the transducer being angled upward to visualize the heart . The mitral valve leaflets were visualized and the transducer then angled until the longest axis of the left ventricle was visualized that clearly demonstrated endocardial echoes around the entire apex . During the study the transducer was angled slightly from side to side to ensure that the longest axis of the left ventricle was visualized . The echocardiograms were recorded during quiet respiration or held partial inspiration (for 1 or s as closely as possible to atmospheric intrathoracic pressures) both at rest and at peak exercise . The Valsalva and Mueller maneuvers were carefully avoided . Calculation of left ventricular volume . End-diastolic (onset of the ECG QRS complex) and end-systolic (the videotape frame just before mitral valve opening) images were identified by off-line analysis with use of forward and reverse slow motion and frame by frame analysis of the videotape . The endocardium was defined as the interface between cavity and myocardial echoes and epicardium was defined as the middle of the brightest pericardial echoes . Subcostal end-diastolic and end-systolic frames that demonstrated the longest possible long axis and in which all or nearly all of the endocardium and epicardium could be visualized on a single stop frame image were identified for each subject at rest, every 3 min during exercise and at peak exercise . The end-diastolic endocardial and epicardial borders and end-systolic endocardial border were traced manually into a microcomputer (Dextra D-100) (including the papillary muscle within the ventricular cavity) with use of an XY digitizer (0 .01 in . [0 .03 cm] resolution) so that the drawn
366
JACC Vol . 14, No. August 1989 :364-71
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
border was directly overlaid on the video image . Left ventricular volume and mass were calculated with use of a single plane Simpson's rule algorithm . This algorithm has been validated by previous investigators (18) in comparison with contrast cineangiography and in this laboratory (17) by comparison of rest and maximal exercise left ventricular ejection fraction in normal humans and patients with coronary artery disease with first pass radionuclide angiography . Ventricular volumes were normalized to body surface area to account for differences in body size . To detect images that did not pass through similar portions of the left ventricle at rest and during exercise, left ventricular mass was calculated at rest and at each stage of exercise . We assumed that the subject's left ventricular mass did not change from rest to exercise, so that any change in calculated left ventricular mass that exceeded the 95% confidence limits of the measurement established by this laboratory using the equipment and methods of analysis employed in the present study (approximately 8%) would be due to foreshortening of the left ventricle . Therefore, any cardiac cycle in which the calculated left ventricular mass differed from the rest left ventricular mass by >10% was excluded from analysis, and another cardiac cycle was chosen for calculation of left ventricular volumes . The rest cardiac cycle was always calculated first to establish the baseline left ventricular mass . Thereafter, cardiac cycles during exercise and at peak exercise for that subject were analyzed in random sequence . Statistical methods . All data are expressed as a mean value ± 1 SD . The significance of rest to peak exercise changes in heart rate, systolic blood pressure, ejection fraction and exercise time was tested by analysis of variance . Differences between the two groups and differences within each group between rest, intermediate and peak exercise left ventricular volumes were tested for significance with use of repeated measures analysis of variance for two factors . A p value <0 .05 was considered statistically significant .
Results Heart rate and systolic blood pressure . For purposes of comparison, each subject's heart rate was normalized to the percent change between rest and peak exercise . Therefore, 0% represented the heart rate at rest, 50% represented the halfway point between the rest and peak exercise heart rate and 100% represented the peak heart rate for that individual . The difference between the athletes and sedentary subjects in rest heart rate (6 ± 10 versus 69 ± 8 beats/min) approached, but did not attain statistical significance (p = 0 .065) . The heart rates for both groups increased progressively during exercise . The heart rate (in beats/min) at 5% (93 ± 10 versus 96 ± 14), 50% (134 ± 14 versus 1 7 ± 16) and 75% (166 ± 10 versus 161 ± 18) of the way to peak exercise
1 0-
100
N
8o
P< .001
p< .05
L
E
E
Athletes
60 #M
M
+
0
w
40
Sedentary
p< .0 meant 1 S .D . Y
0
0
P< .0
* p< .00+ } athletes vs sedentary ** D< .01 + p< .001 vs rest
I I 1 1 1 0 5 50 75 100 PERCENT HEART RATE CHANGE
Figure 1 . Change in end-diastolic volume index (EDVI) during exercise in the athletes and sedentary subjects . In the athletes, end-diastolic volume index increases early in exercise, but enddiastolic volume index decreases as peak exercise is approached . Although the athletes have a larger rest end-diastolic volume index, both groups have approximately parallel changes during exercise and at peak exercise .
and at peak exercise (187 ± 7 versus 186 ± 13) was the same in both groups . There were also no significant differences between the athletes and sedentary subjects at peak exercise in systolic blood pressure (187 ± 16 versus 181 ± 19 mm Hg) or two-dimensional echocardiographic left ventricular ejection fraction (0 .80 ± 0 .08 versus 0 .78 ± 0 .08) . As expected, the athletes had a significantly longer exercise time (16 .3 .6 min) than that of the sedentary controls (1 .0 ± .6 min) (p < 0 .001) . Changes in end-diastolic volume index (Fig . 1). At rest, the athletes' end-diastolic volume index was 85 ± 14 ml/m , increased early in exercise (95 ± 17 ml/m at 5% [p < 0 .001 versus rest], then decreased progressively to 90 ± 17 ml/m at 50%, 83 ± 18 ml/m at 75% of the way to peak heart rate, and at peak exercise it was 61 ± 14 MI/M (p < 0 .001 compared with rest) (Fig . 1) . The sedentary group's enddiastolic volume index was 6 ± 14 ml/m at rest, 67 ± 14 ml/m at 5%, 70 ± 15 ml/m at 50% and 63 ± 15 ml/m at 75% of the way from rest to peak heart rate, and decreased to 46 ± 10 ml/m (p < 0 .001 compared with rest) at peak exercise . The difference in end-diastolic volume index between the groups was significant at rest, during exercise and at peak exercise (p < 0 .01 to 0 .001) (Fig . 1) . Changes in end-systolic volume index (Fig . ) . The athletes' end-systolic volume index at rest was 37 ± 11 ml/m and decreased throughout exercise to 11 ± ml/m at peak exercise (p < 0 .001 compared with rest) . The sedentary
JACC Vol . 14 . No . August 1989 :364-71
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
* p< .001 athletes vs sedentary vs rest vs rest
80
+ p< .001 ++ p< .005
aE
60 p< .05
4
mean ± 1 S .D .
40
W
0 + Athletes ++ Sedentary
0 I 1 I 0 5 50
.01 P< I I 75 100
PERCENT HEART RATE CHANGE Figure . Changes in end-systolic volume index (ESVI) during exercise . The athletes have a larger end-systolic volume index at rest than that of the sedentary subjects, but not during exercise or at peak exercise, and both groups demonstrate a significant decrease in end-systolic volume index at peak exercise .
MI/M subjects' end-systolic volume index was 1 ± 6 at rest MI/M and also decreased during exercise to 10 ± 4 at peak exercise (p < 0 .005 compared with rest) . The differences in end-systolic volume index between the athletes and sedentary group were significant at rest (p < 0 .001) . but not during exercise or at peak exercise . Changes in stroke-volume index (Fig . 3) . At rest, the athletes' stroke volume index was 48 ± 15 ml/m . This value increased significantly (p < 0 .001) at 5% of peak heart rate
Figure 3. Changes in stroke volume index (SVI) during exercise . In the athletes, the stroke volume index increases early in exercise but decreases toward rest volumes at peak exercise, reflecting the reduction in end-diastolic volume index at peak exercise . Both groups demonstrate approximately parallel changes in stroke volume index during exercise .
100
4
I
60
**
Athl e t e s **
40
Sedentary
U) v p< .0
0 *P< .001 **p< .o1
0
and stroke volume index (Fig . 5) at that time . In both the athletes and sedentary subjects, the end-diastolic volume index and stroke volume index tended to decrease as the maximal heart rate was approached . Thirteen of the 15 athletes demonstrated a decrease in end-diastolic volume index and 11 a decrease in stroke volume index when their heart rate exceeded 170 beats/min . Similarly, of the 13 sedentary subjects who achieved a maximal heart rate > 170 beats/min, 9 exhibited a decreased end-diastolic volume index and 9 a decreased stroke volume index at maximal exercise . To summarize, the athletes had significantly larger enddiastolic and end-systolic volume indexes than those of the sedentary adults at rest, but both groups responded with similar directional and percent changes in ventricular volumes during exercise .
Discussion
P< .001
E 7
reflecting the similar magnitude reductions of end-diastolic and end-systolic volume indexes at maximal exercise . The athletes' stroke volume index was significantly larger than that of the sedentary group during exercise (p < 0 .01 to 0 .001) and at peak exercise (p < 0 .01) . Left ventricular volumes related to heart rate (Fig . 4 and 5) . To assess the importance of tachycardia (and shortened diastolic filling time) on left ventricular volume during exercise, comparison was made between heart rate at the end of each 3 min of exercise to end-diastolic volume index (Fig . 4)
r
N
E
to 66 ± 13 ml/m', did not change at 50 and 75% of peak exercise levels and decreased to 49 + 14 ml/m at peak exercise (p < 0 .001 versus 75% level) . The peak exercise stroke volume index was not significantly different from rest because both the end-diastolic volume index and endsystolic volume indexes decreased a similar amount at peak exercise . The sedentary group's stroke volume index was 40 ± 10 ml/m at rest, tended to increase but not significantly (p = 0 .09 to 0 .3 ) at 5, 50 and 75% of the way to peak exercise and was 36 ± 9 MI/M at peak exercise, also
moon ± I S .D. P<.001
80
367
athletes vs sedentary
I I I F I 5 50 75 100 0
PERCENT HEART RATE CHANGE
For years there has been interest in cardiac adaptations to exercise training . Insights into the neurohumoral and hemodynamic mechanisms that control left ventricular function can be gained by understanding the changes that occur in response to the loads applied to the heart during athletic training . As quantitative noninvasive methods for assessing left ventricular performance during stress become more widely used, knowledge of the deviations from expected responses due to an active lifestyle and to disease will be extremely important in the accurate identification of cardiac dysfunction . In addition, by elucidating the normal mechanisms for adapting to aerobic stress, the mechanisms of compensations that occur during pathologic pressure or volume loading states can be better understood .
GINZTON ET AL, EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
368
JACC Vol . 14, No . August 1989 :364-71
140
N
E
1 0
E x
100
0 z W
Figure 4 . End-diastolic volume index versus heart rate, measured every 3 min during exercise and at peak exercise . The initial increase in end-diastolic volume index exhibited by most athletes and sedentary subjects at low levels of exercise is reversed at higher heart rates, and there is a marked reduction in end-diastolic volume index in most subjects whose heart rate exceeded 170 beats/min .
80
f J
...
•
60 J 0
N
40
z
0
a c W
0 40
ATHLETES (n-15) o-c SEDENTARY (n=I4) 1 I I I I I 1 60 80 100 1 0 140 160 180
1 1 00 0
HEART RATE (beats/min)
Previous studies on stroke volumes during exercise . Some previous investigators have observed, using gas exchange kinetics (19), fluoroscopy (14), radionuclide angiography (16, 0, 1, ) and echocardiography (4,9, 3, 4) that in both normal subjects and athletes the stroke volume increases during dynamic exercise . They have concluded, therefore, that the Frank-Starling mechanism is an important adaptation to increasing cardiac output during exercise . Other investigators (5,10), however, have reported that, although stroke volume increases early in the course of dynamic exercise, it does not continue to increase as the intensity of exercise increases and approaches maximal exercise . These authors have concluded that enhanced contractility and tachycardia, rather than the Frank-Starling mechanism, are the primary mechanisms for increasing cardiac output during maximal upright exercise . Most of the previous studies have been required by
s- - • 100-
0 40
technical constraints to evaluate cardiac performance at low or moderate exercise levels (9), to average many cardiac cycles over several minutes (15,16, 1, ) or to perform supine exercise (9,15) . However, these limitations may significantly affect the results of exercise testing . Low level exercise is performed under different left ventricular loading conditions and levels of inotropic stimulation from those of maximal exercise ( 1) . The differences between values for end-diastolic and end-systolic volume indexes at low and moderate levels of exercise and values of maximal exercise observed in this and previous studies in both athletes and sedentary subjects illustrate the differences in left ventricular volumes that these different loading conditions can cause . Averaging many cardiac cycles has been shown to result in erroneous estimation of ejection fraction in normal subjects nearing maximal exercise ( 5) . Lastly, because of the altered preload, maximal supine exercise in normal subjects has
ATHLETES (n •1 5)
o-b SEDENTARY (n-14)
Figure 5. Stroke volume index versus heart rate, measured every 3 min during exercise and at peak exercise . The reduced end-diastolic volume index at high heart rates is accompanied by a reduction in stroke volume index toward rest values in both the athletes and the sedentary control subjects .
I I 1 I I 1 t 60 80 100 1 0 140 160 180 HEART RATE (beats/min)
JACC Vol . 14, No . August 1989 :364-71
been shown (16) to result in different changes in end-diastolic volume and stroke volume from those induced by upright exercise in the same normal individuals . Therefore, data gathered from low level and from supine exercise and time-averaged data may not be directly applicable to assessing left ventricular performance at the peak of maximal upright dynamic exercise . Present study using echocardiography . Two-dimensional echocardiography has the advantages of providing information on left ventricular volume on a heat by beat basis and does not require the averaging of data over many cardiac cycles . We have developed and validated the technique of subcostal view two-dimensional echocardiography during symptom-limited upright bicycle exercise to measure exercise-induced changes in left ventricular volume and function (17) . Therefore, without the technical restrictions on exercise intensity, data collection time and body position, the technique seemed well suited to reexamining changes in left ventricular volume during exercise from onset to maximal effort . In this study, the athletes had a significantly larger rest end-diastolic volume index, end-systolic volume index and stroke volume index than those of the normal subjects . During exercise, both groups had similar degrees of change in ventricular volumes . The end-diastolic volume index decreased at high levels of exercise and at peak exercise it was > 6% less than control values for both the athletes and sedentary subjects . In both groups, end-systolic volume index decreased at moderate and high levels of exercise, reflecting increased left ventricular contractility with exercise . The changes in stroke volume index were also parallel, so that the stroke volume index did not increase proportionately more during exercise in the athletes . Therefore, these data suggest that endurance-trained athletes, although starting from a higher baseline end-diastolic volume index, use mechanisms similar to those of sedentary persons to increase cardiac output during maximal upright dynamic exercise . Comparison with previous studies using radionuclide angiography . Several recent studies have investigated left ventricular volume changes during maximal upright exercise in athletes and sedentary adults . Plotnick et al . ( ) used gated radionuclide angiography to measure left ventricular performance in sedentary subjects and noted that end-diastolic volume index tended to decrease toward control values as the subjects approached peak exercise . These authors did not report the reduction in end-diastolic volume index at peak exercise that was found in the present study . However, the mean peak exercise heart rate in their study was 166 beats/min, whereas in the present study it was 187 beats/min . The influence of very high heart rates on end-diastolic volume index and stroke volume index is readily apparent from Figures 4 and 5, which show that the majority of athletes and sedentary subjects who exceeded a heart rate of
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
369
170 beats/min had further reductions in end-diastolic volume index and stroke volume index as they reached peak exercise . In addition, because of the .5 min data acquisition time required for the radionuclide angiograms in the study of Plotnick et al . ( ), the volume data they reported represent time-averaged values effectively combining intermediate intensity and peak intensity exercise . Rerych et al . ( 0) used first transit radionuclide angiography to measure left ventricular volumes during upright bicycle exercise in a group of competitive collegiate swimmers . They reported significant increases in end-diastolic volume and stroke volume at peak exercise compared with rest, which were not found in the present study . It is possible that the disparate results are due to the different athlete populations studied (swimmers train in the supine position, runners in the upright position) . However, more recent investigators ( 1) have pointed out that the method used by Rerych et al . ( 0) to calculate absolute left ventricular volumes has a large standard error and thus may not be optimal for calculating volumes in this setting . although ejection fraction measurements are highly accurate . The left ventricular volume changes reported in the present study for both the athletes and the sedentary subjects are similar to those of Higgenbotham et al . ( 1), who also studied healthy sedentary adults during maximal upright bicycle exercise with gated radionuclide angiography and hemodynamic monitoring . Although the peak exercise enddiastolic volume index in their study was significantly greater than the control value, they also noted a tendency for the end-diastolic volume index to decrease as the subjects approached peak exercise . In their study, the decreasing end-diastolic volume index near maximal exercise was accompanied by a small but significant increase in pulmonary capillary wedge pressure, suggesting that reduced diastolic filling time due to tachycardia was responsible for the decreased end-diastolic volume index at maximal exercise . The data of the present study are consistent with this conclusion because both end-diastolic volume index and stroke volume index tended to decrease rapidly in both the athletes and sedentary subjects when their heart rate exceeded 170 beats/min (Fig . 4 and 5) . A recent study ( 6) using two-dimensional echocardiography, thermodilution and Fick oximetry to measure stroke volume during exercise in normal subjects also demonstrated that stroke volume increased early in upright exercise, but did not increase further to maximal exercise . Comparison with previous studies using echocardiography . In an echocardiographic study of competitive and noncompetitive long distance runners, Crawford et al . (4) reported that elite runners increased cardiac output by progressively increasing stroke volume more than did the nonelite athletes, who increased cardiac output by increasing contractility more than did the elite runners . These results were not observed in the present study . However, both groups of
370
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
athletes studied by Crawford et al . (4), especially the noncompetitive runners, had a lower maximal exercise heart rate than those of our study subjects so that direct comparison of their maximal exercise left ventricular volumes with the volumes observed at peak exercise in the present study may not be appropriate . The peak exercise heart rates in the study of Crawford et al . are comparable with the heart rates at the 75% level in the present study, at which time the end-diastolic volume index and stroke volume index had not decreased below rest values . Therefore, when left ventricular volume responses in these studies are compared by heart rate attained, the data are in agreement that moderate intensity exercise results in increased end-diastolic volume index and stroke volume index compared with rest . Mechanism of adaptation to exercise . The data reported here are in agreement with most previous studies (4,10, 6) that end-diastolic volume index and stroke volume index increase during low level and moderate upright bicycle exercise in healthy adults and highly trained athletes . The finding that end-diastolic volume index and stroke volume index decrease significantly at peak exercise compared with rest in both athletes and sedentary subjects has not previously been reported, and suggests that at near maximally attainable heart rate, the reduced diastolic filling described by Higginbotham et al . ( 1) may be a major mechanism for limiting stroke volume index and, thereby, cardiac output . Therefore, the data of the present and previous studies indicate that the healthy left ventricle has a bimodal mechanism of adapting to upright bicycle exercise . At low and moderate exercise, end-diastolic and stroke volume indexes increase and end-systolic volume decreases . At high intensity exercise, however, the effect of tachycardia on diastolic time and filling limits end-diastolic volume and, therefore, stroke volume drops . Therefore, at peak exercise, particularly at heart rates greater than approximately 170 beats/min, maximal cardiac output is attained by increased contractility and tachycardia, and the Frank-Starling mechanism no longer is a factor . Critique of methodology. In this study, maximal oxygen uptake was not measured directly during exercise, so it is not possible to state categorically that either the athletes or the sedentary subjects reached maximal exercise . However, all subjects exercised to the point of exhaustion . In addition, the heart rate achieved during exercise by the athletes was 93 ± 3% of age-predicted maximal heart rate and that achieved by the sedentary subjects was 96 ± 5% of that rate . Two-dimensional echocardiography is a tomographic imaging technique and considerable care must be taken to ensure that consistent orientation of the echocardiographic transducer in relation to the heart and similar long-axis views of the heart can be obtained at rest and at exercise . Measuring left ventricular mass to identify and reject those cardiac cycles that do not include consistent views of the left ventricle is an important step in assuring the accuracy and
JACC Vol . 14, No. August 1989 :364-71
reproducibility of the left ventricular volume measurements in this study (although <10% of the cardiac cycles were rejected and another selected by this criterion .) We have previously validated (17) the subcostal view echocardiographic measurements of left ventricular ejection fraction against first pass radionuclide angiography at rest and at peak exercise over a wide range of ejection fraction in normal subjects and patients with coronary artery disease ; we observed a high correlation (r = 0 .91, p < 0 .001) between the two techniques . In addition, we have conducted studies ( 7) of segmental wall motion in normal subjects and patients with a previous myocardial infarction and have also shown significant correlation with segmental wall motion measured by first pass radionuclide angiography at rest and at exercise . The accuracy and reproducibility of the single plane Simpson's rule volume algorithm used here has also been validated in this laboratory by comparing two-dimensional echocardiographic measurements of left ventricular mass in dogs with autopsy left ventricular weight (r = 0 .86, p < 0 .001, SEE = 11 .9 g) and has been validated by others (18) by comparison with contrast cineangiography . The reproducibility of two-dimensional echocardiographic measurements of left ventricular volumes in humans ( 8) was found to be high as long as the same observer made the sequential measurements (a procedure adhered to in this study) . The ventricular volumes and stroke volumes reported in this study are slightly smaller than those previously obtained with gas exchange kinetic ( 9), indicator-dilution (19, 1) and radionuclide angiographic ( 1, ) techniques . This difference is at least in part due to systematic underestimation of ventricular volumes by the echocardiographic equipment used . This underestimation of cavity area, however, is linear over a wide range of values, with a significant correlation with actual cavity area (30) and volume (18, 6) . Because the purpose of this study was to compare relative left ventricular volume changes in athletes and sedentary subjects using the same equipment and exercise conditions, we chose to report unaltered left ventricular volume data . Conclusion . Long-term competitive highly trained long distance runners have larger end-diastolic volumes and end-systolic volumes at rest compared with those of normal sedentary adults and the differences in end-diastolic volume and stroke volume persist (but do not increase) during moderate and maximal upright bicycle exercise . The rest left ventricular dilation in the athletes suggests that they may use the Frank-Starling mechanism at rest to maintain a normal cardiac output at rest and to increase cardiac output during low and moderate intensity exercise . However, during maximal exercise, the end-diastolic volume is reduced and stroke volume is unchanged in both the athletes and the sedentary group . Therefore, enhanced contractility and tachycardia, and not the Frank-Starling mechanism, appear to be the major mechanisms for increasing cardiac output during maximal
JACC Vol . 14, No . August 1989 :364-71
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
upright dynamic exercise in highly trained endurance athletes and healthy sedentary adults . Although aerobically trained long distance runners have a larger left ventricle than that of sedentary subjects at rest, both groups use similar bimodal mechanisms for increasing cardiac output during maximal upright bicycle exercise .
References
371
volume relationship : a sensitive indicator of left ventricular dysfunction in patients with coronary artery disease . Circulation 1981 ;63 :1008-18 . 16. Poliner L, Dehmer G, Lewis S, Parkey R, Blomqvist C, Willerson J . Left ventricular performance in normal subjects : a comparison of the responses to exercise in the upright and supine positions . Circulation 1980 :6 :5 8-34 . 17 . Ginzton L, Laks M, Brizendine M, Conant R, Mena 1 . Non-invasive measurement of the rest and exercise peak systolic pressure/end systolic volume ratio : a sensitive two-dimensional echocardiographic indicator of left ventricular function . J Am Coll Cardiol 1984 ;4 :509-16 .
I . Fagard R, Aubert A, Lysens R, Staessen J, Vanhees L . Amery A . Noninvasive assessment of seasonal variations in cardiac structure and function in cyclists . Circulation 1983 ;67 :896-901 .
18 . Erbel R, Schweizer P, Meyer J, Grenner H, Krebs W, Effert S . Left ventricular volume and ejection fraction determined by cross-sectional echocardiography in patients with coronary artery disease : a prospective study . Clin Cardiol 1980 ;3 :377-83 .
. DeMaria AN, Neumann A, Lee G, Fowler W, Mason DT . Alterations in ventricular mass and performance induced by exercise training in man evaluated by echocardiography. Circulation 1978 ;57 : 37-44 .
19 . Ekblom B, Astrand P, Saltin B, Stenberg J, Wallstrom B . Effect of training on circulatory response to exercise . J Appi Physiol 1968 ; 4 :518- 8 .
3 . Ehsani AA, Hagberg JM, Hickson RC . Rapid changes in left ventricular dimensions and mass in response to physical conditioning and deconditioning . Am J Cardiol 1978 ;4 :5 -6.
0 . Rerych S, Scholz P, Sabiston D, Jones R . Effects of exercise training on left ventricular function in normal subjects : a longitudinal study by radionuclide angiography . Am J Cardiol 1980 ;45 : 44-5 .
4 . Crawford MH . Petru MA . Rabinowitz C . Effect of isotonic exercise training on left ventricular volume during upright exercise . Circulation 1986;7 :1 37-43 .
1 . Higginbotham MB, Morris KG, Williams RS, McHale PA, Coleman RE, Cobb FR . Regulation of stroke volume during submaximal and maximal upright exercise in normal man . Circ Res 1986;58 : 81-91 .
5 . Ekelund L . Circulatory and respiratory adaptation during prolonged exercise . Acta Physiol Scand 1967 ;70(suppl 9 ) :1-38 .
. Plotnick GD, Becker LC, Fisher ML, et al . Use of the Frank-Starling mechanism during submaximal versus maximal upright exercise . Am J Physiol 1986 : 51 (Heart Circ Physiol 0) :HII01-H5 .
6 . Longhurst JC, Kelly AR, Gonyea WJ, Mitchell JH . Chronic training with static and dynamic exercise : cardiovascular adaptation, and response to exercise . Circ Res 1981 ;48(suppl1) :1-171-8 . 7 . Huston TP, Puffer JC, Rodney WM . The athletic heart syndrome . N Engl J Med 1985 ;313 : 4-3 . 8 . Colan SD, Sanders SP, MacPherson D, Borow KM . Left ventricular diastolic function in elite athletes with physiologic cardiac hypertrophy . J Am Coll Cardiol 1985 ;6:545-9 . 9 . Weiss J, Weisfeldt M, Mason S, Garrison J, Livengood S, Fortuin N . Evidence of Frank-Starling effect in man during severe semisupine exercise . Circulation 1979 ;59 :655-61 . 10 . Rost R, Hollmann W . Athlete's heart-a review of its historical assessment and new aspects . Int J Sports Med 1983 ;4 :147-65 . 11 . Badeer H . The genesis of cardiomegaly in strenuous athletic training : a new look . J Sports Med Phys Fitness 1975 ;15 :57-67 .
3 . Crawford M, Amon K, Vance W . Exercise -dimensional echocardiography . Quantitation of left ventricular performance in patients with severe angina pectoris . Am J Cardiol 1983 ;51 :1-6 . 4 . Dickhuth H, Simon G, Heiss H, Lehmann M, Wybitul K, Keul J . Comparative echocardiographic examinations in sitting and supine position at rest and during dynamic exercise . Int J Sports Med 1981 : :178-81 . 5 . Crawford M, Petru M, Amon K . Sorensen S, Vanre W . Comparative value of -dimensional echocardiography and radionuclide angiography for quantitating changes in left ventricular performance during exercise limited by angina pectoris . Am J Cardiol 1984;53 :4 -6 . 6 . Christie J, Sheldahl LM, Tristani FE, Sagar KB, Ptacin MJ, Wann S . Determination of stroke volume and cardiac output during exercise : comparison of two-dimensional and Doppler echocardiography, Fick oximetry, and thermodilution . Circulation 1987 ;76 :539-47.
1 . Gilbert C, Nutter D, Felner J, Perkins J, Heymsfield S, Schlant R . Echocardiographic study of cardiac dimensions and function in the endurance-trained athlete . Am J Cardiol 1977 ;40 :5 8-33 .
7 . Ginzton L, Conant R, Brizendine M, Lee F, Mena 1, Laks M . Exercise subcostal two-dimensional echocardiography : a new method of segmental wall motion analysis . Am J Cardiol 1984 ;53 :805-11 .
13 . Nishimura T . Yamada Y, Kawai C . Echocardiographic evaluation of long-term effects of exercise on left ventricular hypertrophy and function in professional bicyclists . Circulation 1980 ;61 :83 -40 .
8 . Gordon EP, Schnittger I, Fitzgerald PJ, Williams P, Popp RL . Reproducibility of left ventricular volumes by two-dimensional echocardiography . J Am Coll Cardiol 1983 ; :506-13 .
14 . Sharma B, Goodwin J, Raphael M, Steiner R, Rainbow R, Taylor S . Left ventricular angiography on exercise : a new method of assessing left ventricular function in ischaemic heart disease . Br Heart J 1976 ;38 :59-70 .
9. Astrand P, Cuddy T, Saltin B . Stenberg J . Cardiac output during submaximal and maximal work . J Appl Physiol 1964 ;19: 68-74 .
15 . Dehmer G, Lewis S, Hillis D, Corbett J, Parkey R, Willerson J . Exercise-induced alterations in left ventricular volumes and the pressure-
30. Helak J, Reichek N . Quantitation of human left ventricular mass and volume by two-dimensional echocardiography : in vitro anatomic validation . Circulation 1981 ;63 :1398-407 .
364
Effect of Long-Term High Intensity Aerobic Training on Left Ventricular Volume During Maximal Upright Exercise LEONARD E . GINZTON, MD, FACC, RICHARD CONANT, PHD, MARIANNE BRIZENDINE, RN, MICHAEL M . LAKS, MD, FACC Torrance, California
The purpose of this study was to determine whether high intensity, long-term aerobic training causes the left ventricle to develop different mechanisms for increasing cardiac output during submaximal and maximal upright bicycle exercise. Fifteen competitive collegiate long distance runners and 14 healthy sedentary adults were studied with use of subcostal view four chamber two-dimensional echocardiography at rest and during and at peak maximal upright bicycle exercise . At rest, the athletes had a larger end-diastolic volume index (85 ± 14 ml/m ) ( mean ± 1 SD) than that of the sedentary adults (6 ± 14 MI/M ) and a larger end-systolic volume index (37 ± 11 versus 1 ± 6 MI/M ) . During low and moderate intensity exercise, end-diastolic and stroke volume indexes increased in both groups, but at high intensity exercise and at peak exercise the end-diastolic volume index of both groups decreased significantly below
Athletes engaged in long-term high intensity aerobic training are known to have a dilated left ventricle at rest (1-4) . However, the mechanisms of the left ventricular dilation, and the possibility that the mechanisms for increasing cardiac output during exercise differ in sedentary persons and athletes, have long been a subject of dispute (4-8) . Elucidation of these mechanisms would have implications both for exercise training and for understanding the role of exercise in myocardial hypertrophy and adaptation to left ventricular volume overload states . Some investigators (7,9) have reported that an increased cardiac output during exercise is achieved by increasing left ventricular end-diastolic volume
From the Division of Cardiology, Department of Medicine, HarborUniversity of California at Los Angeles Medical Center and the Department of Medicine, University of California at Los Angeles School of Medicine, Torrance, California . Manuscript received August 5, 1988 : revised manuscript received February 15, 1989, accepted March 3, 1989 . Address for reprints : Leonard E . Ginzton, MD, Division of Cardiology, Harbor-University of California at Los Angeles Medical Center, 1000 West Carson Street, Torrance, California 90509 . (0 1989 by the American College of Cardiology
rest value (athletes, 61 ± 14 ; sedentary subjects, 46 ± 10 mI/mz , both p < 0 .001 compared with rest) . Reflecting the decreased end-diastolic volume index, at peak exercise, the stroke volume index had decreased from intermediate exercise values in both groups and was not different from rest values. Therefore, although long distance runners have a dilated left ventricle at rest, they utilize the same mechanisms as sedentary adults for increasing cardiac output during upright dynamic exercise . At low and moderate level exercise, the Frank-Starling mechanism is a dominant mechanism for increasing cardiac output, but at peak exercise, probably because of reduced diastolic left ventricular filling, enhanced contractility is the major mechanism for maintaining stroke volume . (J Am Coll Cardiol 1989 ;14:364-71)
and stroke volume, especially in highly trained endurance athletes (4) . These data have been cited as evidence that the Frank-Starling mechanism is utilized preferentially by athletes during exercise and that athletes and sedentary subjects, therefore, have fundamentally different mechanisms for increasing cardiac output during exercise . However, other investigators (5,10) have reported that stroke volume increases early in exercise in highly trained athletes but does not continue to increase with further increases in exercise load to maximal exercise . These investigators have concluded that increased contractility, rather than the FrankStarling mechanism, is the predominant mechanism by which cardiac output is increased during exercise in athletes . They have proposed that the left ventricular dilation observed in athletes may be secondary to a bradycardia at rest that causes the left ventricle to dilate to maintain a normal cardiac output at rest (10-13) . In many previous studies of cardiovascular dynamics during exercise in athletes, exercise testing has often been carried out in the supine position (9,14,15), which poses 0735-1097/89/$3 .50
JACC Vol . 14, No . August 1989 :364-71
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
different preload and afterload conditions on the heart from those posed by upright exercise . Also, left ventricular volume estimates have often been made by techniques that require up to several minutes of data gathering, which may obscure the effects of rapidly changing left ventricular loading conditions and function, especially at maximal exercise (15,16) . Therefore, it has been difficult to apply these data meaningfully to maximal exercise in the upright position . Our laboratory has reported (17) that subcostal view two-dimensional echocardiography is a useful, accurate method for obtaining left ventricular volume measurements on a beat by beat basis during and at the peak of maximal upright bicycle exercise . Therefore, this technique seemed applicable to the study of changes in left ventricular volume in endurance-trained athletes . We have observed that in sedentary, healthy adults, the end-diastolic volume decreases at the peak of symptom-limited upright bicycle exercise and that, at peak exercise, stroke volume is unchanged from control . We hypothesized that endurance athletes would have a similar response during exercise . If the athletes' end-diastolic volume decreases at peak exercise, this would suggest that their increased exercise cardiac output is primarily related to increased heart rate and myocardial contractility, and that they do not utilize the Frank-Starling mechanism to adapt to maximal upright exercise . Therefore, the purpose of this study was to compare changes in left ventricular volume during maximal upright bicycle exercise in competitive long distance runners and sedentary normal adults .
Methods Study groups . Two groups of subjects were studied . The athletic group was made up of 15 competitive long distance runners, all of whom were members of their collegiate cross-country running team . Their age was 19 ± 1 years (mean ± I SD) ; 9 were male and 6 female . The runners had averaged training distances of 116 ± 5 km/week for an average of 4 .4 ± 1 years . All had exceeded 70 km/week in training for at least the previous 8 consecutive weeks . The control group comprised 14 sedentary, healthy adults whose mean age was 8 ± 6 years ; 10 were male and 4 female . None of the control group had participated in any competitive or regular conditioning activity for >5 years . None of the subjects in either group had any clinical or historical evidence of cardiovascular disease or hypertension . None was taking any medication and all had a normal cardiac physical examination and a normal 1 lead electrocardiogram (ECG) both at rest and at the peak of maximal upright bicycle exercise . All subjects were studied according to a research protocol approved by the Research Committee and Human Subjects Committee of Harbor-University of California at Los Ange-
365
les Medical Center and all subjects gave written informed consent before participating in the study . Exercise protocol . All subjects were studied at the same time of day (mid afternoon) . They were exercised in the upright position on a bicycle ergometer with continuous ECG monitoring . A 1 lead ECG and sphygmomanometric blood pressure were obtained at rest, every 3 min during exercise and at peak exercise . Exercise was begun at a zero work load and increased by 75 kilopond meters (kp-m)/min each minute until the subject adamantly refused further exercise because of exhaustion . Echocardiographic protocol . Subcostal four chamber view two-dimensional echocardiograms were obtained with use of a wide angle mechanical scanner (Advanced Technologies Laboratories Mark III) . The subcostal four chamber view was recorded on videotape with the subject sitting upright at rest, every 3 min during exercise and from approximately 15 s before the attainment of peak exercise until peak exercise was reached . No post-exercise images were used for measurement of left ventricular volumes . To obtain reproducible views of the left ventricle, the following, previously reported protocol (17) was employed . The handheld transducer was placed in the subxiphoid position just below or to the left of the xiphoid process and gently pushed inward and upward with the transducer being angled upward to visualize the heart . The mitral valve leaflets were visualized and the transducer then angled until the longest axis of the left ventricle was visualized that clearly demonstrated endocardial echoes around the entire apex . During the study the transducer was angled slightly from side to side to ensure that the longest axis of the left ventricle was visualized . The echocardiograms were recorded during quiet respiration or held partial inspiration (for 1 or s as closely as possible to atmospheric intrathoracic pressures) both at rest and at peak exercise . The Valsalva and Mueller maneuvers were carefully avoided . Calculation of left ventricular volume . End-diastolic (onset of the ECG QRS complex) and end-systolic (the videotape frame just before mitral valve opening) images were identified by off-line analysis with use of forward and reverse slow motion and frame by frame analysis of the videotape . The endocardium was defined as the interface between cavity and myocardial echoes and epicardium was defined as the middle of the brightest pericardial echoes . Subcostal end-diastolic and end-systolic frames that demonstrated the longest possible long axis and in which all or nearly all of the endocardium and epicardium could be visualized on a single stop frame image were identified for each subject at rest, every 3 min during exercise and at peak exercise . The end-diastolic endocardial and epicardial borders and end-systolic endocardial border were traced manually into a microcomputer (Dextra D-100) (including the papillary muscle within the ventricular cavity) with use of an XY digitizer (0 .01 in . [0 .03 cm] resolution) so that the drawn
366
JACC Vol . 14, No. August 1989 :364-71
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
border was directly overlaid on the video image . Left ventricular volume and mass were calculated with use of a single plane Simpson's rule algorithm . This algorithm has been validated by previous investigators (18) in comparison with contrast cineangiography and in this laboratory (17) by comparison of rest and maximal exercise left ventricular ejection fraction in normal humans and patients with coronary artery disease with first pass radionuclide angiography . Ventricular volumes were normalized to body surface area to account for differences in body size . To detect images that did not pass through similar portions of the left ventricle at rest and during exercise, left ventricular mass was calculated at rest and at each stage of exercise . We assumed that the subject's left ventricular mass did not change from rest to exercise, so that any change in calculated left ventricular mass that exceeded the 95% confidence limits of the measurement established by this laboratory using the equipment and methods of analysis employed in the present study (approximately 8%) would be due to foreshortening of the left ventricle . Therefore, any cardiac cycle in which the calculated left ventricular mass differed from the rest left ventricular mass by >10% was excluded from analysis, and another cardiac cycle was chosen for calculation of left ventricular volumes . The rest cardiac cycle was always calculated first to establish the baseline left ventricular mass . Thereafter, cardiac cycles during exercise and at peak exercise for that subject were analyzed in random sequence . Statistical methods . All data are expressed as a mean value ± 1 SD . The significance of rest to peak exercise changes in heart rate, systolic blood pressure, ejection fraction and exercise time was tested by analysis of variance . Differences between the two groups and differences within each group between rest, intermediate and peak exercise left ventricular volumes were tested for significance with use of repeated measures analysis of variance for two factors . A p value <0 .05 was considered statistically significant .
Results Heart rate and systolic blood pressure . For purposes of comparison, each subject's heart rate was normalized to the percent change between rest and peak exercise . Therefore, 0% represented the heart rate at rest, 50% represented the halfway point between the rest and peak exercise heart rate and 100% represented the peak heart rate for that individual . The difference between the athletes and sedentary subjects in rest heart rate (6 ± 10 versus 69 ± 8 beats/min) approached, but did not attain statistical significance (p = 0 .065) . The heart rates for both groups increased progressively during exercise . The heart rate (in beats/min) at 5% (93 ± 10 versus 96 ± 14), 50% (134 ± 14 versus 1 7 ± 16) and 75% (166 ± 10 versus 161 ± 18) of the way to peak exercise
1 0-
100
N
8o
P< .001
p< .05
L
E
E
Athletes
60 #M
M
+
0
w
40
Sedentary
p< .0 meant 1 S .D . Y
0
0
P< .0
* p< .00+ } athletes vs sedentary ** D< .01 + p< .001 vs rest
I I 1 1 1 0 5 50 75 100 PERCENT HEART RATE CHANGE
Figure 1 . Change in end-diastolic volume index (EDVI) during exercise in the athletes and sedentary subjects . In the athletes, end-diastolic volume index increases early in exercise, but enddiastolic volume index decreases as peak exercise is approached . Although the athletes have a larger rest end-diastolic volume index, both groups have approximately parallel changes during exercise and at peak exercise .
and at peak exercise (187 ± 7 versus 186 ± 13) was the same in both groups . There were also no significant differences between the athletes and sedentary subjects at peak exercise in systolic blood pressure (187 ± 16 versus 181 ± 19 mm Hg) or two-dimensional echocardiographic left ventricular ejection fraction (0 .80 ± 0 .08 versus 0 .78 ± 0 .08) . As expected, the athletes had a significantly longer exercise time (16 .3 .6 min) than that of the sedentary controls (1 .0 ± .6 min) (p < 0 .001) . Changes in end-diastolic volume index (Fig . 1). At rest, the athletes' end-diastolic volume index was 85 ± 14 ml/m , increased early in exercise (95 ± 17 ml/m at 5% [p < 0 .001 versus rest], then decreased progressively to 90 ± 17 ml/m at 50%, 83 ± 18 ml/m at 75% of the way to peak heart rate, and at peak exercise it was 61 ± 14 MI/M (p < 0 .001 compared with rest) (Fig . 1) . The sedentary group's enddiastolic volume index was 6 ± 14 ml/m at rest, 67 ± 14 ml/m at 5%, 70 ± 15 ml/m at 50% and 63 ± 15 ml/m at 75% of the way from rest to peak heart rate, and decreased to 46 ± 10 ml/m (p < 0 .001 compared with rest) at peak exercise . The difference in end-diastolic volume index between the groups was significant at rest, during exercise and at peak exercise (p < 0 .01 to 0 .001) (Fig . 1) . Changes in end-systolic volume index (Fig . ) . The athletes' end-systolic volume index at rest was 37 ± 11 ml/m and decreased throughout exercise to 11 ± ml/m at peak exercise (p < 0 .001 compared with rest) . The sedentary
JACC Vol . 14 . No . August 1989 :364-71
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
* p< .001 athletes vs sedentary vs rest vs rest
80
+ p< .001 ++ p< .005
aE
60 p< .05
4
mean ± 1 S .D .
40
W
0 + Athletes ++ Sedentary
0 I 1 I 0 5 50
.01 P< I I 75 100
PERCENT HEART RATE CHANGE Figure . Changes in end-systolic volume index (ESVI) during exercise . The athletes have a larger end-systolic volume index at rest than that of the sedentary subjects, but not during exercise or at peak exercise, and both groups demonstrate a significant decrease in end-systolic volume index at peak exercise .
MI/M subjects' end-systolic volume index was 1 ± 6 at rest MI/M and also decreased during exercise to 10 ± 4 at peak exercise (p < 0 .005 compared with rest) . The differences in end-systolic volume index between the athletes and sedentary group were significant at rest (p < 0 .001) . but not during exercise or at peak exercise . Changes in stroke-volume index (Fig . 3) . At rest, the athletes' stroke volume index was 48 ± 15 ml/m . This value increased significantly (p < 0 .001) at 5% of peak heart rate
Figure 3. Changes in stroke volume index (SVI) during exercise . In the athletes, the stroke volume index increases early in exercise but decreases toward rest volumes at peak exercise, reflecting the reduction in end-diastolic volume index at peak exercise . Both groups demonstrate approximately parallel changes in stroke volume index during exercise .
100
4
I
60
**
Athl e t e s **
40
Sedentary
U) v p< .0
0 *P< .001 **p< .o1
0
and stroke volume index (Fig . 5) at that time . In both the athletes and sedentary subjects, the end-diastolic volume index and stroke volume index tended to decrease as the maximal heart rate was approached . Thirteen of the 15 athletes demonstrated a decrease in end-diastolic volume index and 11 a decrease in stroke volume index when their heart rate exceeded 170 beats/min . Similarly, of the 13 sedentary subjects who achieved a maximal heart rate > 170 beats/min, 9 exhibited a decreased end-diastolic volume index and 9 a decreased stroke volume index at maximal exercise . To summarize, the athletes had significantly larger enddiastolic and end-systolic volume indexes than those of the sedentary adults at rest, but both groups responded with similar directional and percent changes in ventricular volumes during exercise .
Discussion
P< .001
E 7
reflecting the similar magnitude reductions of end-diastolic and end-systolic volume indexes at maximal exercise . The athletes' stroke volume index was significantly larger than that of the sedentary group during exercise (p < 0 .01 to 0 .001) and at peak exercise (p < 0 .01) . Left ventricular volumes related to heart rate (Fig . 4 and 5) . To assess the importance of tachycardia (and shortened diastolic filling time) on left ventricular volume during exercise, comparison was made between heart rate at the end of each 3 min of exercise to end-diastolic volume index (Fig . 4)
r
N
E
to 66 ± 13 ml/m', did not change at 50 and 75% of peak exercise levels and decreased to 49 + 14 ml/m at peak exercise (p < 0 .001 versus 75% level) . The peak exercise stroke volume index was not significantly different from rest because both the end-diastolic volume index and endsystolic volume indexes decreased a similar amount at peak exercise . The sedentary group's stroke volume index was 40 ± 10 ml/m at rest, tended to increase but not significantly (p = 0 .09 to 0 .3 ) at 5, 50 and 75% of the way to peak exercise and was 36 ± 9 MI/M at peak exercise, also
moon ± I S .D. P<.001
80
367
athletes vs sedentary
I I I F I 5 50 75 100 0
PERCENT HEART RATE CHANGE
For years there has been interest in cardiac adaptations to exercise training . Insights into the neurohumoral and hemodynamic mechanisms that control left ventricular function can be gained by understanding the changes that occur in response to the loads applied to the heart during athletic training . As quantitative noninvasive methods for assessing left ventricular performance during stress become more widely used, knowledge of the deviations from expected responses due to an active lifestyle and to disease will be extremely important in the accurate identification of cardiac dysfunction . In addition, by elucidating the normal mechanisms for adapting to aerobic stress, the mechanisms of compensations that occur during pathologic pressure or volume loading states can be better understood .
GINZTON ET AL, EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
368
JACC Vol . 14, No . August 1989 :364-71
140
N
E
1 0
E x
100
0 z W
Figure 4 . End-diastolic volume index versus heart rate, measured every 3 min during exercise and at peak exercise . The initial increase in end-diastolic volume index exhibited by most athletes and sedentary subjects at low levels of exercise is reversed at higher heart rates, and there is a marked reduction in end-diastolic volume index in most subjects whose heart rate exceeded 170 beats/min .
80
f J
...
•
60 J 0
N
40
z
0
a c W
0 40
ATHLETES (n-15) o-c SEDENTARY (n=I4) 1 I I I I I 1 60 80 100 1 0 140 160 180
1 1 00 0
HEART RATE (beats/min)
Previous studies on stroke volumes during exercise . Some previous investigators have observed, using gas exchange kinetics (19), fluoroscopy (14), radionuclide angiography (16, 0, 1, ) and echocardiography (4,9, 3, 4) that in both normal subjects and athletes the stroke volume increases during dynamic exercise . They have concluded, therefore, that the Frank-Starling mechanism is an important adaptation to increasing cardiac output during exercise . Other investigators (5,10), however, have reported that, although stroke volume increases early in the course of dynamic exercise, it does not continue to increase as the intensity of exercise increases and approaches maximal exercise . These authors have concluded that enhanced contractility and tachycardia, rather than the Frank-Starling mechanism, are the primary mechanisms for increasing cardiac output during maximal upright exercise . Most of the previous studies have been required by
s- - • 100-
0 40
technical constraints to evaluate cardiac performance at low or moderate exercise levels (9), to average many cardiac cycles over several minutes (15,16, 1, ) or to perform supine exercise (9,15) . However, these limitations may significantly affect the results of exercise testing . Low level exercise is performed under different left ventricular loading conditions and levels of inotropic stimulation from those of maximal exercise ( 1) . The differences between values for end-diastolic and end-systolic volume indexes at low and moderate levels of exercise and values of maximal exercise observed in this and previous studies in both athletes and sedentary subjects illustrate the differences in left ventricular volumes that these different loading conditions can cause . Averaging many cardiac cycles has been shown to result in erroneous estimation of ejection fraction in normal subjects nearing maximal exercise ( 5) . Lastly, because of the altered preload, maximal supine exercise in normal subjects has
ATHLETES (n •1 5)
o-b SEDENTARY (n-14)
Figure 5. Stroke volume index versus heart rate, measured every 3 min during exercise and at peak exercise . The reduced end-diastolic volume index at high heart rates is accompanied by a reduction in stroke volume index toward rest values in both the athletes and the sedentary control subjects .
I I 1 I I 1 t 60 80 100 1 0 140 160 180 HEART RATE (beats/min)
JACC Vol . 14, No . August 1989 :364-71
been shown (16) to result in different changes in end-diastolic volume and stroke volume from those induced by upright exercise in the same normal individuals . Therefore, data gathered from low level and from supine exercise and time-averaged data may not be directly applicable to assessing left ventricular performance at the peak of maximal upright dynamic exercise . Present study using echocardiography . Two-dimensional echocardiography has the advantages of providing information on left ventricular volume on a heat by beat basis and does not require the averaging of data over many cardiac cycles . We have developed and validated the technique of subcostal view two-dimensional echocardiography during symptom-limited upright bicycle exercise to measure exercise-induced changes in left ventricular volume and function (17) . Therefore, without the technical restrictions on exercise intensity, data collection time and body position, the technique seemed well suited to reexamining changes in left ventricular volume during exercise from onset to maximal effort . In this study, the athletes had a significantly larger rest end-diastolic volume index, end-systolic volume index and stroke volume index than those of the normal subjects . During exercise, both groups had similar degrees of change in ventricular volumes . The end-diastolic volume index decreased at high levels of exercise and at peak exercise it was > 6% less than control values for both the athletes and sedentary subjects . In both groups, end-systolic volume index decreased at moderate and high levels of exercise, reflecting increased left ventricular contractility with exercise . The changes in stroke volume index were also parallel, so that the stroke volume index did not increase proportionately more during exercise in the athletes . Therefore, these data suggest that endurance-trained athletes, although starting from a higher baseline end-diastolic volume index, use mechanisms similar to those of sedentary persons to increase cardiac output during maximal upright dynamic exercise . Comparison with previous studies using radionuclide angiography . Several recent studies have investigated left ventricular volume changes during maximal upright exercise in athletes and sedentary adults . Plotnick et al . ( ) used gated radionuclide angiography to measure left ventricular performance in sedentary subjects and noted that end-diastolic volume index tended to decrease toward control values as the subjects approached peak exercise . These authors did not report the reduction in end-diastolic volume index at peak exercise that was found in the present study . However, the mean peak exercise heart rate in their study was 166 beats/min, whereas in the present study it was 187 beats/min . The influence of very high heart rates on end-diastolic volume index and stroke volume index is readily apparent from Figures 4 and 5, which show that the majority of athletes and sedentary subjects who exceeded a heart rate of
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
369
170 beats/min had further reductions in end-diastolic volume index and stroke volume index as they reached peak exercise . In addition, because of the .5 min data acquisition time required for the radionuclide angiograms in the study of Plotnick et al . ( ), the volume data they reported represent time-averaged values effectively combining intermediate intensity and peak intensity exercise . Rerych et al . ( 0) used first transit radionuclide angiography to measure left ventricular volumes during upright bicycle exercise in a group of competitive collegiate swimmers . They reported significant increases in end-diastolic volume and stroke volume at peak exercise compared with rest, which were not found in the present study . It is possible that the disparate results are due to the different athlete populations studied (swimmers train in the supine position, runners in the upright position) . However, more recent investigators ( 1) have pointed out that the method used by Rerych et al . ( 0) to calculate absolute left ventricular volumes has a large standard error and thus may not be optimal for calculating volumes in this setting . although ejection fraction measurements are highly accurate . The left ventricular volume changes reported in the present study for both the athletes and the sedentary subjects are similar to those of Higgenbotham et al . ( 1), who also studied healthy sedentary adults during maximal upright bicycle exercise with gated radionuclide angiography and hemodynamic monitoring . Although the peak exercise enddiastolic volume index in their study was significantly greater than the control value, they also noted a tendency for the end-diastolic volume index to decrease as the subjects approached peak exercise . In their study, the decreasing end-diastolic volume index near maximal exercise was accompanied by a small but significant increase in pulmonary capillary wedge pressure, suggesting that reduced diastolic filling time due to tachycardia was responsible for the decreased end-diastolic volume index at maximal exercise . The data of the present study are consistent with this conclusion because both end-diastolic volume index and stroke volume index tended to decrease rapidly in both the athletes and sedentary subjects when their heart rate exceeded 170 beats/min (Fig . 4 and 5) . A recent study ( 6) using two-dimensional echocardiography, thermodilution and Fick oximetry to measure stroke volume during exercise in normal subjects also demonstrated that stroke volume increased early in upright exercise, but did not increase further to maximal exercise . Comparison with previous studies using echocardiography . In an echocardiographic study of competitive and noncompetitive long distance runners, Crawford et al . (4) reported that elite runners increased cardiac output by progressively increasing stroke volume more than did the nonelite athletes, who increased cardiac output by increasing contractility more than did the elite runners . These results were not observed in the present study . However, both groups of
370
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
athletes studied by Crawford et al . (4), especially the noncompetitive runners, had a lower maximal exercise heart rate than those of our study subjects so that direct comparison of their maximal exercise left ventricular volumes with the volumes observed at peak exercise in the present study may not be appropriate . The peak exercise heart rates in the study of Crawford et al . are comparable with the heart rates at the 75% level in the present study, at which time the end-diastolic volume index and stroke volume index had not decreased below rest values . Therefore, when left ventricular volume responses in these studies are compared by heart rate attained, the data are in agreement that moderate intensity exercise results in increased end-diastolic volume index and stroke volume index compared with rest . Mechanism of adaptation to exercise . The data reported here are in agreement with most previous studies (4,10, 6) that end-diastolic volume index and stroke volume index increase during low level and moderate upright bicycle exercise in healthy adults and highly trained athletes . The finding that end-diastolic volume index and stroke volume index decrease significantly at peak exercise compared with rest in both athletes and sedentary subjects has not previously been reported, and suggests that at near maximally attainable heart rate, the reduced diastolic filling described by Higginbotham et al . ( 1) may be a major mechanism for limiting stroke volume index and, thereby, cardiac output . Therefore, the data of the present and previous studies indicate that the healthy left ventricle has a bimodal mechanism of adapting to upright bicycle exercise . At low and moderate exercise, end-diastolic and stroke volume indexes increase and end-systolic volume decreases . At high intensity exercise, however, the effect of tachycardia on diastolic time and filling limits end-diastolic volume and, therefore, stroke volume drops . Therefore, at peak exercise, particularly at heart rates greater than approximately 170 beats/min, maximal cardiac output is attained by increased contractility and tachycardia, and the Frank-Starling mechanism no longer is a factor . Critique of methodology. In this study, maximal oxygen uptake was not measured directly during exercise, so it is not possible to state categorically that either the athletes or the sedentary subjects reached maximal exercise . However, all subjects exercised to the point of exhaustion . In addition, the heart rate achieved during exercise by the athletes was 93 ± 3% of age-predicted maximal heart rate and that achieved by the sedentary subjects was 96 ± 5% of that rate . Two-dimensional echocardiography is a tomographic imaging technique and considerable care must be taken to ensure that consistent orientation of the echocardiographic transducer in relation to the heart and similar long-axis views of the heart can be obtained at rest and at exercise . Measuring left ventricular mass to identify and reject those cardiac cycles that do not include consistent views of the left ventricle is an important step in assuring the accuracy and
JACC Vol . 14, No. August 1989 :364-71
reproducibility of the left ventricular volume measurements in this study (although <10% of the cardiac cycles were rejected and another selected by this criterion .) We have previously validated (17) the subcostal view echocardiographic measurements of left ventricular ejection fraction against first pass radionuclide angiography at rest and at peak exercise over a wide range of ejection fraction in normal subjects and patients with coronary artery disease ; we observed a high correlation (r = 0 .91, p < 0 .001) between the two techniques . In addition, we have conducted studies ( 7) of segmental wall motion in normal subjects and patients with a previous myocardial infarction and have also shown significant correlation with segmental wall motion measured by first pass radionuclide angiography at rest and at exercise . The accuracy and reproducibility of the single plane Simpson's rule volume algorithm used here has also been validated in this laboratory by comparing two-dimensional echocardiographic measurements of left ventricular mass in dogs with autopsy left ventricular weight (r = 0 .86, p < 0 .001, SEE = 11 .9 g) and has been validated by others (18) by comparison with contrast cineangiography . The reproducibility of two-dimensional echocardiographic measurements of left ventricular volumes in humans ( 8) was found to be high as long as the same observer made the sequential measurements (a procedure adhered to in this study) . The ventricular volumes and stroke volumes reported in this study are slightly smaller than those previously obtained with gas exchange kinetic ( 9), indicator-dilution (19, 1) and radionuclide angiographic ( 1, ) techniques . This difference is at least in part due to systematic underestimation of ventricular volumes by the echocardiographic equipment used . This underestimation of cavity area, however, is linear over a wide range of values, with a significant correlation with actual cavity area (30) and volume (18, 6) . Because the purpose of this study was to compare relative left ventricular volume changes in athletes and sedentary subjects using the same equipment and exercise conditions, we chose to report unaltered left ventricular volume data . Conclusion . Long-term competitive highly trained long distance runners have larger end-diastolic volumes and end-systolic volumes at rest compared with those of normal sedentary adults and the differences in end-diastolic volume and stroke volume persist (but do not increase) during moderate and maximal upright bicycle exercise . The rest left ventricular dilation in the athletes suggests that they may use the Frank-Starling mechanism at rest to maintain a normal cardiac output at rest and to increase cardiac output during low and moderate intensity exercise . However, during maximal exercise, the end-diastolic volume is reduced and stroke volume is unchanged in both the athletes and the sedentary group . Therefore, enhanced contractility and tachycardia, and not the Frank-Starling mechanism, appear to be the major mechanisms for increasing cardiac output during maximal
JACC Vol . 14, No . August 1989 :364-71
GINZTON ET AL . EXERCISE LEFT VENTRICULAR VOLUME IN ATHLETES
upright dynamic exercise in highly trained endurance athletes and healthy sedentary adults . Although aerobically trained long distance runners have a larger left ventricle than that of sedentary subjects at rest, both groups use similar bimodal mechanisms for increasing cardiac output during maximal upright bicycle exercise .
References
371
volume relationship : a sensitive indicator of left ventricular dysfunction in patients with coronary artery disease . Circulation 1981 ;63 :1008-18 . 16. Poliner L, Dehmer G, Lewis S, Parkey R, Blomqvist C, Willerson J . Left ventricular performance in normal subjects : a comparison of the responses to exercise in the upright and supine positions . Circulation 1980 :6 :5 8-34 . 17 . Ginzton L, Laks M, Brizendine M, Conant R, Mena 1 . Non-invasive measurement of the rest and exercise peak systolic pressure/end systolic volume ratio : a sensitive two-dimensional echocardiographic indicator of left ventricular function . J Am Coll Cardiol 1984 ;4 :509-16 .
I . Fagard R, Aubert A, Lysens R, Staessen J, Vanhees L . Amery A . Noninvasive assessment of seasonal variations in cardiac structure and function in cyclists . Circulation 1983 ;67 :896-901 .
18 . Erbel R, Schweizer P, Meyer J, Grenner H, Krebs W, Effert S . Left ventricular volume and ejection fraction determined by cross-sectional echocardiography in patients with coronary artery disease : a prospective study . Clin Cardiol 1980 ;3 :377-83 .
. DeMaria AN, Neumann A, Lee G, Fowler W, Mason DT . Alterations in ventricular mass and performance induced by exercise training in man evaluated by echocardiography. Circulation 1978 ;57 : 37-44 .
19 . Ekblom B, Astrand P, Saltin B, Stenberg J, Wallstrom B . Effect of training on circulatory response to exercise . J Appi Physiol 1968 ; 4 :518- 8 .
3 . Ehsani AA, Hagberg JM, Hickson RC . Rapid changes in left ventricular dimensions and mass in response to physical conditioning and deconditioning . Am J Cardiol 1978 ;4 :5 -6.
0 . Rerych S, Scholz P, Sabiston D, Jones R . Effects of exercise training on left ventricular function in normal subjects : a longitudinal study by radionuclide angiography . Am J Cardiol 1980 ;45 : 44-5 .
4 . Crawford MH . Petru MA . Rabinowitz C . Effect of isotonic exercise training on left ventricular volume during upright exercise . Circulation 1986;7 :1 37-43 .
1 . Higginbotham MB, Morris KG, Williams RS, McHale PA, Coleman RE, Cobb FR . Regulation of stroke volume during submaximal and maximal upright exercise in normal man . Circ Res 1986;58 : 81-91 .
5 . Ekelund L . Circulatory and respiratory adaptation during prolonged exercise . Acta Physiol Scand 1967 ;70(suppl 9 ) :1-38 .
. Plotnick GD, Becker LC, Fisher ML, et al . Use of the Frank-Starling mechanism during submaximal versus maximal upright exercise . Am J Physiol 1986 : 51 (Heart Circ Physiol 0) :HII01-H5 .
6 . Longhurst JC, Kelly AR, Gonyea WJ, Mitchell JH . Chronic training with static and dynamic exercise : cardiovascular adaptation, and response to exercise . Circ Res 1981 ;48(suppl1) :1-171-8 . 7 . Huston TP, Puffer JC, Rodney WM . The athletic heart syndrome . N Engl J Med 1985 ;313 : 4-3 . 8 . Colan SD, Sanders SP, MacPherson D, Borow KM . Left ventricular diastolic function in elite athletes with physiologic cardiac hypertrophy . J Am Coll Cardiol 1985 ;6:545-9 . 9 . Weiss J, Weisfeldt M, Mason S, Garrison J, Livengood S, Fortuin N . Evidence of Frank-Starling effect in man during severe semisupine exercise . Circulation 1979 ;59 :655-61 . 10 . Rost R, Hollmann W . Athlete's heart-a review of its historical assessment and new aspects . Int J Sports Med 1983 ;4 :147-65 . 11 . Badeer H . The genesis of cardiomegaly in strenuous athletic training : a new look . J Sports Med Phys Fitness 1975 ;15 :57-67 .
3 . Crawford M, Amon K, Vance W . Exercise -dimensional echocardiography . Quantitation of left ventricular performance in patients with severe angina pectoris . Am J Cardiol 1983 ;51 :1-6 . 4 . Dickhuth H, Simon G, Heiss H, Lehmann M, Wybitul K, Keul J . Comparative echocardiographic examinations in sitting and supine position at rest and during dynamic exercise . Int J Sports Med 1981 : :178-81 . 5 . Crawford M, Petru M, Amon K . Sorensen S, Vanre W . Comparative value of -dimensional echocardiography and radionuclide angiography for quantitating changes in left ventricular performance during exercise limited by angina pectoris . Am J Cardiol 1984;53 :4 -6 . 6 . Christie J, Sheldahl LM, Tristani FE, Sagar KB, Ptacin MJ, Wann S . Determination of stroke volume and cardiac output during exercise : comparison of two-dimensional and Doppler echocardiography, Fick oximetry, and thermodilution . Circulation 1987 ;76 :539-47.
1 . Gilbert C, Nutter D, Felner J, Perkins J, Heymsfield S, Schlant R . Echocardiographic study of cardiac dimensions and function in the endurance-trained athlete . Am J Cardiol 1977 ;40 :5 8-33 .
7 . Ginzton L, Conant R, Brizendine M, Lee F, Mena 1, Laks M . Exercise subcostal two-dimensional echocardiography : a new method of segmental wall motion analysis . Am J Cardiol 1984 ;53 :805-11 .
13 . Nishimura T . Yamada Y, Kawai C . Echocardiographic evaluation of long-term effects of exercise on left ventricular hypertrophy and function in professional bicyclists . Circulation 1980 ;61 :83 -40 .
8 . Gordon EP, Schnittger I, Fitzgerald PJ, Williams P, Popp RL . Reproducibility of left ventricular volumes by two-dimensional echocardiography . J Am Coll Cardiol 1983 ; :506-13 .
14 . Sharma B, Goodwin J, Raphael M, Steiner R, Rainbow R, Taylor S . Left ventricular angiography on exercise : a new method of assessing left ventricular function in ischaemic heart disease . Br Heart J 1976 ;38 :59-70 .
9. Astrand P, Cuddy T, Saltin B . Stenberg J . Cardiac output during submaximal and maximal work . J Appl Physiol 1964 ;19: 68-74 .
15 . Dehmer G, Lewis S, Hillis D, Corbett J, Parkey R, Willerson J . Exercise-induced alterations in left ventricular volumes and the pressure-
30. Helak J, Reichek N . Quantitation of human left ventricular mass and volume by two-dimensional echocardiography : in vitro anatomic validation . Circulation 1981 ;63 :1398-407 .