Systemic arterial compliance (C) and vascular resistance (R) regulate effective arterial elastance (Ea), an index of artery load. Increases in Ea during exercise are due primarily to reductions of C and maintain optimal ventricular-arterial coupling. Because C at rest and left ventricular functional reserve are greater in endurance-trained (ET) compared with sedentary control (SC) humans, we hypothesized that reductions of C and increases in Ea are greater in ET than SC individuals. The aim of this study was to investigate C, R, and Ea during exercise in ET and SC humans. C, R, Ea, and cardiac cycle length (T) were measured at rest and during exercise of 40, 60, and 80% maximal oxygen uptake using Doppler ultrasonography in 12 SC and 13 ET men. C decreased in an exercise intensity-dependent manner in both groups, but its reductions were greater in the ET than SC subjects. Consequently, although C at rest was greater in the ET than SC group, the intergroup difference in C disappeared during exercise. Exercise-related changes in R/T were relatively slight and R/T was lower in the ET than the SC group, both at rest and during exercise. Although Ea at rest was lower in the ET than SC group, there were no intergroup differences in Ea at 40, 60, or 80% maximal oxygen uptake. We conclude that the reductions of C from rest to exercise are more marked in ET than SC humans. This may be related to the exercise-associated disappearance of the difference in Ea between ET and SC humans.
- Doppler ultrasonography
- endurance exercise training
effective arterial elastance (Ea), an index of arterial load (14, 41, 42), increases during exercise and maintains optimal ventricular-arterial coupling (20, 26, 43). Consequently, the arterial system can achieve efficacious and maximal stroke work from the left ventricle during exercise and can respond to peripheral demand with a minimum of energy consumption by the left ventricle (20, 43). Cross-sectional and intervention studies have reported that endurance exercise training increases cardiac systolic function reserve (33, 38). Seals et al. (33) have demonstrated that body surface area-corrected cardiac output (CO) during exercise was greater in endurance-trained (ET) athletes compared with sedentary control (SC) men, although it did not differ under the resting condition between the groups. They have also suggested that changes in ejection fraction from rest to exercise were more marked in the ET humans than the SC humans (33). Stratton et al. (38) have showed that 6-mo endurance training program (walking, jogging, or bicycling; 45 min/session; 4–5 sessions/wk) increased CO, normalized by body surface area, and ejection fraction during exercise without the increases in these measures at rest. To respond to these remarkable increases in cardiac systolic function and maintain optimal ventricular-arterial coupling, increases in Ea during exercise may also be greater in ET compared with SC humans. However, it remains unclear whether increases in Ea during exercise are greater in ET humans compared with SC individuals.
Ea is primarily regulated by systemic vascular resistance (R) and systemic arterial compliance (C). It is well known that the combination of the R-to-cardiac cycle length (T) ratio (R/T), and C explains the variability seen in Ea (7, 28, 34). Although the contribution of R/T to Ea is greater under resting conditions (7, 34), we have demonstrated that the increases in Ea during exercise are driven mainly by reductions of C (28). If the increases in Ea are more marked in ET than SC humans, the reductions of C would be, at least partly, associated with this phenomenon. C is reduced by the increased heart rate (HR) (19, 23) or blood pressure (6). HR in ET humans is lower than SC humans at rest but is comparable to that in SC humans at maximal exercise load, suggesting that HR reserve is greater in ET relative to SC individuals (33, 38). Although there are no differences in blood pressure at rest between the ET and SC humans, blood pressure is higher at maximal exercise load in the ET subjects, indicating the increases in blood pressure from rest to maximal exercise are greater in the ET than SC humans (33, 38). Additionally, increased arterial compliance at rest in ET humans (4, 6, 12, 15, 29, 30, 32, 40, 45–47), compared with SC humans, may increase the C reserve during exercise. Taken together, it is possible that the reductions of C from rest to exercise are more remarkable in ET humans compared with SC humans. However, it remains unclear whether the reductions of C during exercise in ET men are greater than in SC men and whether the increases in Ea during exercise are greater in ET compared with SC individuals.
We hypothesized that the reductions of C and increases in Ea during exercise are greater in ET than SC humans. We also predicted that the more marked reductions of C in ET compared with SC humans are associated with the greater increases in Ea during exercise in the ET humans. The aim of this study was to investigate C, R/T, and Ea during exercise in ET and age-matched SC men. As in our previous study, we tested this hypothesis by calculating C, R/T, and Ea from arterial pressure and blood flow and T measured by Doppler ultrasonography at rest and during semirecumbent cycling exercise at work rates corresponding to 40%, 60%, and 80% maximal oxygen uptake (28).
Male long- or middle-distance runners (ET; n = 13) and age-matched SC men (n = 12) volunteered to participate in this study. The ET men were intercollegiate athletes belonging to track and field teams, and their competitive sports careers were longer than 2 years. Their training regimen included 5.6 ± 0.2 sessions/wk averaging 2.4 ± 0.2 h/session. Athletes who had been concurrently performing both regular endurance and strength training (i.e., cross-training) were excluded, because C might be affected differently by endurance, strength, and cross trainings (13, 25, 27, 29, 30, 46). The SC men had sedentary lifestyles (no regular physical activity) for at least 2 yr. All subjects were free of signs, symptoms, and history of any overt chronic diseases. None of the participants had a history of smoking, and none were currently taking any medications.
The present study was approved by the Ethical Committee of the Institute of Health and Sport Sciences of the University of Tsukuba. This study conformed to the principles outlined in the Helsinki Declaration. All subjects gave their written informed consent before inclusion into this study.
This study was designed according to our previous study (28). The subjects were examined on two occasions, at least 48 h apart but within 7 days. Before all tests, subjects refrained from alcohol consumption and intense physical activity (exercise) for 24 h and caffeine consumption for 4 h to avoid immediate (acute) physiological effects. All measurements were taken in a quiet, temperature-controlled room (25°C) after a resting period of at least 20 min. First, maximal oxygen uptake was measured (AE280S; Minato Medical Science, Osaka, Japan). On the second day of testing, 8 min of exercise were sequentially performed at each exercise intensity (i.e., 40%, 60%, and 80% maximal oxygen uptake) in a semisupine position on a cycle ergometer (232CXL; Combi Wellness, Tokyo, Japan) fitted with a backrest. Breath-by-breath oxygen uptake was measured using an online computer-assisted system (AE280S; Minato Medical Science), and work rate was controlled to maintain the target exercise intensity. Blood pressure, stroke volume (SV), and HR were measured at rest and during exercise.
Maximal oxygen uptake.
Maximal oxygen uptake was determined using incremental cycling to exhaustion (3 min at 80 W, with a 30-W increase every 3 min) by monitoring breath-by-breath oxygen uptake and carbon dioxide production (AE280S; Minato Medical Science), HR, and ratings of perceived exertion.
Two-dimensional echocardiography and Doppler ultrasonography.
SV at rest and during exercise was measured using a Doppler-ultrasonographic system (EnVisor; Koninklijke Philips Electronics, Eindhoven, The Netherlands), as previously described (28). Briefly, the insertion point of the aortic valve tips at end-diastole at rest and during exercise of 40, 60, and 80% maximal oxygen uptake was defined by two-dimensional imaging in the parasternal long axis view with a 3.5-MHz transducer, and the M-mode echocardiogram was stored. Doppler ultrasonographic flow velocity curves in the ascending aorta were continuously recorded using a 1.9-MHz probe held in the suprasternal notch. SV was calculated as a product of the velocity-time integral and the aortic cross-sectional area (Image J; National Institutes of Health, Bethesda, MD). The left ventricular morphology and function (i.e., ejection fraction and fractional shortening) at rest were also measured before exercise sessions using the M-mode echocardiogram in the long axis view (28–30).
Blood pressure, used to calculate C, R, and Ea, was measured using cuff sphygmomanometry according to previous studies (9, 26, 28). Systolic and diastolic blood pressure (SBP and DBP, respectively) in the semisupine position on the ergometer was recorded three times at rest and during steady-state exercise. These measurements were taken using a mercury manometer, a standard adult Velcro cuff system with rubber bladder, and a stethoscope. Pulse pressure (PP) and mean blood pressure (MBP) were calculated as follows: PP = SBP − DBP and MBP = PP/3+DBP.
Blood pressure, measured using cuff sphygmomanometry, and SV were used to calculate C, R, and Ea at rest and during exercise (9, 26, 28). These indices were calculated as follows (7, 9, 14, 28): Ea = 0.9 × SBP/SV, C = SV/PP, and R = MBP/(SV × HR) To compare the relationships of Ea (mmHg/ml) to C (ml/mmHg) and R (mmHg·s·ml−1), R was corrected by T (R/T, mmHg/ml) (7, 28, 34).
Data are expressed as means (SD). To compare the hemodynamics at rest and during exercise between the ET and SC groups, statistical analysis was carried out using two-way repeated measures ANOVA (group × exercise). In the case of significant F values, the unpaired t-test was used to determine the effects of the group at respective conditions, and the post hoc test (Fisher's protected least significant difference) was used to identify the effect of exercise. P < 0.05 was accepted as significant.
Table 1 shows the subject characteristics of the SC and ET groups. There were no statistical differences in age, height, weight, left ventricular average wall thickness, ejection fraction, or fractional shortening between the groups. Maximal oxygen uptake was statistically greater in the ET than SC subjects. Aortic cross-sectional area, and left ventricular dimensions and mass were also statistically higher in the ET compared with SC men.
Hemodynamics at rest and during exercise are summarized in Table 2. Under resting conditions, there were no statistical differences in blood pressure between the groups. SBP, MBP, and PP statistically increased in an exercise intensity-dependent manner. SBP and PP at 40% and 80% were statistically higher in the ET group compared with the SC group. We did not find statistical effects of group and/or exercise on DBP. HR statistically increased and T statistically decreased in an exercise intensity-dependent manner in both groups. There were no statistical differences in aortic mean and peak blood flow velocities between the groups. These two measures in both groups statistically increased in an exercise intensity-dependent manner. SV was statistically higher at all exercise intensities compared with rest in both groups. R statistically decreased in an exercise intensity-dependent manner except from 60% to 80% maximal oxygen uptake. No statistical differences in R were observed between groups.
Figure 1 shows Ea, C, and R/T at rest and during exercise. Ea did not differ between rest and 40% maximal oxygen uptake (SC, P = 0.18; ET, P = 0.31) but increased at 60% (P = 0.02 and P < 0.01, respectively) and 80% (P < 0.01 and P < 0.01, respectively) compared with baseline. C decreased in an exercise intensity-dependent manner both in the SC (rest vs. 40% maximal oxygen uptake, P < 0.01; vs. 60%, P < 0.01; vs. 80%, P < 0.01) and ET (P < 0.01, P < 0.01, and P < 0.01, respectively) groups. The R/T ratio decreased at 40% maximal oxygen uptake compared with rest, although the decrease in the ET men was not statistically significant (SC, P = 0.01; ET, P = 0.06). It recovered at 60% in the ET group (SC, P = 0.01; ET, P = 0.81) and at 80% in the SC group (SC, P = 0.73; ET, P = 0.21) compared with the baseline. Ea (P < 0.01) and R/T (P < 0.01) at rest were lower, and C (P = 0.02) was higher in the ET group than the SC group. R/T was lower in the ET than the SC group not only at rest but also during exercise (40% maximal oxygen uptake, P = 0.04; 60%, P = 0.12; 80%, P = 0.02). However, the intergroup differences in C disappeared during exercise (P = 0.83, P = 0.37, and P = 0.40, respectively). No differences in Ea between groups were observed during exercise (P = 0.12, P = 0.17, and P = 0.08, respectively).
The changes in Ea, C, and R/T from baseline to the respective exercise intensity are demonstrated in Fig. 2. Two-way ANOVA did not reveal intergroup differences in the changes in Ea from rest to exercise. The reductions of C from rest to exercise were statistically more marked in the ET group compared with the SC group. There were no statistical differences in the changes in R/T from baseline to exercise between groups.
In addition to the relative exercise intensity-matched comparisons, the absolute oxygen uptake-matched comparisons [i.e., the exercise of 60% in the SC men (26.8 ml·kg−1·min−1) vs. that of 40% in the ET men (24.2 ml·kg−1·min−1), comparison 1, and the exercise of 80% in the SC men (35.7 ml·kg−1·min−1) vs. that of 60% in the ET men (36.2 ml·kg−1·min−1], comparison 2) were performed by using unpaired-t test. Table 3 shows the hemodynamics at comparisons 1 and 2. SBP and PP were statistically lower in the ET relative to SC subjects at comparison 1. At comparisons 1 and 2, HR and mean blood flow velocities were statistically lower, and SV and T were statistically higher in the ET than SC humans. The oxygen uptake-matched comparisons of changes in hemodynamics from rest to exercise are summarized in Table 4. The changes in SBP, PP, and peak blood flow velocity at comparison 1 and in HR and mean blood flow velocity at both comparisons were lower in the ET compared with SC group.
Figures 3 and 4 demonstrate associations of absolute oxygen uptake with Ea, C, R/T, and with the changes in these measures from rest to exercise. Ea was lower in the ET than SC group (comparison 1, P < 0.01 and comparison 2, P < 0.01; Fig. 3A), but the changes in Ea from rest to exercise did not differ between groups (P = 0.37 and P = 0.31, respectively; Fig. 4A). C during exercise was greater in the ET group compared with the SC group (P < 0.01 and P < 0.01, respectively; Fig. 3B), but there were no intergroup differences in changes in C from baseline (P = 0.72 and P = 0.24, respectively; Fig. 4B). The ET group showed lower R/T during exercise compared with the SC group (P < 0.01 and P = 0.01, respectively; Fig. 3C), but we did not find the intergroup differences in changes in R/T from baseline (P = 0.68 and P = 0.96, respectively; Fig. 4C).
The salient findings of this study are that the reductions of C from rest to exercise of relatively the same intensities were greater in the ET compared with SC group. Although Ea at rest was lower in the ET relative to SC men, there were no intergroup differences in Ea during the relatively parallel-intensity exercise. At the absolute oxygen uptake-matched comparisons, changes in C and Ea from rest to exercise were comparable between the groups, although the extents of increases in blood pressure and HR were lower in the ET than the SC subjects. We conclude that the greater absolute exercise intensities and the more marked arterial sensitivity to changes in hemodynamics in endurance-trained than sedentary humans appear to induce the greater reductions of arterial compliance during exercise of relatively comparable intensity. It may be associated with the relatively same intensity exercise-associated disappearance of the difference in effective arterial elastance between endurance-trained and sedentary men.
It has been established that endurance exercise training increases arterial compliance. We and other groups have reported that arterial compliance at rest is greater in physically active or ET than SC humans (4, 12, 15, 29, 30, 32, 45, 47). Intervention studies have also demonstrated that endurance exercise training increases arterial compliance (6, 40, 46). The results herein under resting conditions are consistent with these previous studies. On the other hand, the reductions of C during relatively parallel-intensity exercise were greater in the ET compared with SC group. Furthermore, the changes in C from rest to exercise of absolutely the same intensity were comparable between the groups, although the increases in blood pressure and HR were less marked in the ET than the SC group. This phenomenon may be associated with the fact that the reductions of C corresponding to blood pressure increases are greater in compliant arteries than stiff arteries (i.e., compliant arteries are more sensitive to blood pressure changes than stiff arteries) (15). Because reduced arterial compliance under resting conditions is a cardiovascular risk factor (3, 18), the increased arterial compliance achieved by exercise training is implicated in the clinical benefits of exercise training. Additionally, we propose that the greater arterial compliance at rest in endurance-trained compared with sedentary men participates in the increased arterial compliance reserve during exercise.
The greater aerobic capacity in the ET relative to SC subjects results in the relatively similar but absolutely greater exercise intensities. These absolutely increased exercise intensities seem to be associated with more marked blood pressure increases during the exercise: blood pressure during exercise of relatively the same intensities was higher in the ET group than in the SC group, although no intergroup difference was observed at rest. On the other hand, HR at rest was lower in the ET group than the SC group, and this difference remained during relatively the same intensity exercise. Since the increased blood pressure and HR reduce C (6, 19, 23), it would be reasonable to consider that the changes in C during exercise are passively regulated by blood pressure and HR. However, the contribution to the greater reduction of C during relatively the same intensity exercise in the ET compared with SC group may be higher for blood pressure than HR. It is likely that the higher absolute exercise intensities induce more remarkable increases in blood pressure during exercise of relatively comparable intensity and consequently elicit the greater reductions of C in the ET than SC individuals.
Although the reductions of C from rest to relative intensity-matched exercise were greater in the ET than SC group, there were no differences in the extent of changes in Ea between the groups. Ratios of contribution of C and R/T to Ea were 0.45 vs. 1.00 at rest and 0.96 vs. 1.00 at 60% maximal oxygen uptake (28), suggesting that the contribution of C to Ea increases during exercise compared with at rest, but R/T also definitely participates in the regulation of Ea during exercise. In this study, since the change in R/T from rest to exercise was comparable between the groups, it seems that the change in C from rest to exercise of relatively the same intensity could not elicit the clear intergroup differences in the change in Ea.
We calculated C as SV/PP in this study. Although C is the relative changes in arterial volume to changes in blood pressure, not only the time integral of blood flow but also the instantaneous flow can affect SV/PP. The increases in blood flow velocity make the interpretation of comparisons of SV/PP among at rest and respective exercise intensity complicated. If we could perform central arterial waveform analyses to estimate C, the present results would be strengthened. However, there were no intergroup differences in mean and peak blood flow velocities at rest and during relative intensity-matched exercise in coincidence with the previous study (24). Accordingly, we consider that the differences in SV/PP between the groups reflect the differences in C. Also, if tachycardia shortens left ventricular ejection time and reduces SV, SV/PP can underestimate C. However, the intergroup differences in HR at rest between the groups were maintained during exercise of relatively the same intensities. It may be reasonable to consider that there were no marked intergroup differences in the ejection-time changes during the relatively parallel-intensity exercise. At the absolute oxygen uptake-matched comparisons, the changes in SV/PP were comparable between the groups, although the increases in HR were greater in the SC relative to ET group. It is possible that the reductions of actual C during absolutely parallel-intensity exercise were greater in the ET than SC individuals. This speculation does not strongly support our hypothesis but at least does not conflict with the hypothesis that reductions of C during exercise are greater in ET than SC humans.
Blood perfusion during exercise is preferentially distributed to the working limbs and increases in an exercise intensity-dependent manner (5). Blood flow is determined by vascular resistance and the proximal-to-distal pressure gradient. Thus, lower vascular resistance (i.e., higher vascular conductance) would increase peripheral blood perfusion (5). In this study, R/T was lower in the ET group than the SC group both at rest and during exercise. It seems reasonable to conclude that the lower vascular resistance in the ET compared with SC group increases muscular blood flow. On the other hand, it decreases left ventricular afterload and may disturb the balance between ventricular and vascular elastance, which may reduce myocardial energy efficiency (43). However, the greater reductions of C in the ET than SC group result in increased Ea during relative intensity-matched exercise even though R/T was maintained at a lower level. Additionally, the reduction of C-induced increases in central arterial blood pressure may result in the increases in proximal to distal pressure gradient and peripheral blood flow. The coordinated regulation of cardiac and arterial function is essential for efficient aerobic exercise. The greater reduction of arterial compliance and the maintained lower vascular resistance in endurance-trained compared with sedentary men may be reasonable adaptations that allow simultaneous attainment of higher arterial load and peripheral perfusion during aerobic exercise.
Because there were no intergroup differences in changes in Ea, C, and R/T from rest to oxygen uptake-matched exercise, the intergroup differences in these measures during the exercise are likely to reflect the differences at rest. The main determinants of C are arterial volume and stiffness. Aortic cross-sectional area was larger in the ET relative to SC group, in agreement with the previous study (24). Greater aortic volume in the ET than SC subjects may be a causal mechanism responsible for the intergroup differences in C. Arterial stiffness is partly modulated by nitric oxide (NO), a potent vasodilator (16, 39, 48). Endurance exercise increases aortic NO bioavailability (8, 35, 44) and plasma levels of end product of NO (21, 22). The greater C at rest in the ET than SC subjects may also be associated with NO-related reductions of arterial stiffness. R/T is inversely correlated with peripheral vascular cross-sectional area. Endurance exercise increases capillary supply in heart (11) and skeletal muscle (1, 10). This may be an explanation of the lower R/T in the ET compared with the SC group. These factors may be implicated in the intergroup differences in Ea via changes in C and R/T.
SV was greater in the ET than SC group. If arterial wall were more distended by the increased ratio of SV to arterial volume, it might decrease C. Also, increased SV-related increases in peripheral blood flow velocity may increase R/T. However, aortic cross-sectional area was larger in the ET compared with SC group. Again, it has been well known that endurance exercise training increases capillary supply (1, 10, 11). The greater C and lower R/T in the ET subjects with larger SV compared with the SC subjects with smaller SV may suggest that the effects of increased SV in the ET subjects on artery system are counteracted by the enlarged arterial volume.
We used brachial artery blood pressure to calculate C, R/T, and Ea, according to previous studies (9, 26, 28), because it is unable to measure aortic blood pressure noninvasively. Peripheral (e.g., brachial or radial) arterial SBP and PP are higher than central (e.g., aortic) values. These differences enhance with increasing exercise intensity (31, 36). Rowell et al. (31) have demonstrated that the differences in SBP between aorta and radial artery could reach as high as 100 mmHg at maximal exercise, although the differences were smaller at submaximal than maximal exercise and were smaller between aorta and brachial artery than between aorta and radial artery. The degree of the differences is greater in young healthy humans compared with older humans (37). Also, the formula calculating MBP may underestimate actual MBP during exercise because the exercise-related decreases in T primarily depend on left ventricular diastolic time. These are study limitations that should be noted. The next conceivable step is to measure aortic blood pressure and to propose the more precise relationship of C and R/T to Ea in ET humans. However, there were no differences in age between groups in this study. Additionally, Kroeker and Wood (17) have reported that the brachial arterial pressure during exercise as percentages of aortic pressure (SBP, 109%; DBP, 96%) is similar to resting condition (SBP, 111%; DBP, 97%), although we should note that they have not supported the use of peripheral pressure (17). Thus, the tendency for changes in Ea, C, and R/T, calculated from brachial arterial pressure, from rest to exercise, may be similar to those from aortic values, although these measures are overestimated or underestimated. We believe that this study is valid as an initial step to investigate the regulation of arterial compliance and vascular resistance during exercise in ET humans.
Perspectives and Significance
It is not easy to totally elucidate the regulations of cardiovascular system during exercise and its adaptations to exercise training. These issues become more complicated by the presence of different types of exercise. Although aerobic exercise training increases C (4, 6, 12, 15, 29, 30, 32, 40, 45–47), previous studies have suggested that muscle-strengthening exercise training decreases it (2, 13, 25, 29, 30). Aerobic exercise training-related increases in C would be associated with the benefits of aerobic exercise. However, it has been unclear whether the strength exercise training-related reductions of C are unfavored or favored changes. The present findings not only elucidate the reasonable regulation of arterial system during aerobic exercise in the ET humans but also may provide clues to understanding the strength exercise-related arterial stiffening. The results in this study suggest that greater C in the ET compared with SC men augment the extent of C reduction during the relative intensity-matched exercise and may elicit an appropriate central and peripheral blood circulatory condition. It is possible that the changes in arterial properties from rest to strength exercise in strength-trained humans are different from those in SC or ET humans, leading to suitable circulatory condition during the exercise. Future research should focus on systematically elucidating the regulations of arterial system during exercise.
This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (18300215, 18650186, 17700486).
We thank Mr. Keigo Ohyama Byun and Ms. Kayo Morooka for supporting our study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society