Exercise: where the body leads and the heart must follow

At rest and during exercise, cardiac output is regulated to closely match the body's oxygen demand. In resting dogs with impaired electrical conductivity from the SA node to the ventricles (AV block), increasing heart rate by ventricular pacing (from 100 to 180 beats min⁻¹) does not elicit changes in cardiac output or arterial pressure, whereas stroke volume (SV) decreases (White et al. 1971); but it is unknown if humans respond similarly to increases in heart rate due to anatomical and physiological differences (i.e. blood volume distribution). In a recent article in The Journal of Physiology, Bada et al. (2012) explored the effects of heart rate and peripheral vasodilatation on cardiac output with the use of atrial pacing in humans. The authors postulated that increasing heart rate via atrial pacing would attenuate SV due to reductions in central venous and left ventricular filling pressure.

The investigators studied young male subjects during atrial pacing (up to 150 beats min⁻¹) under three conditions (rest, one-leg knee-extensor exercise, and ATP infusion) while measuring central and peripheral haemodynamic parameters. The authors reported that an increased heart rate was accompanied by a proportional decrease in SV and central venous pressure, which maintained cardiac output during all three conditions. Atrial pacing also increased mean arterial pressure at rest and during ATP infusion. These results agree with those seen in dogs and indicate that central venous pressure and peripheral vasodilatation, rather than heart rate, are the primary regulators of cardiac output in humans.

In order to test whether peripheral vasodilatation determines cardiac output, ATP was infused into the femoral artery to elicit vasodilatation and increase blood flow. ATP infusion was adjusted to match the blood flow measured during the single-leg exercise condition. The authors clearly demonstrated that in the absence of exercise, peripheral vasodilatation during atrial pacing maintained cardiac output at a level similar to what was shown during exercise. However, peripheral vasodilatation during exercise is also due to functional sympatholysis and the release of other vasoactive factors. Another striking difference between ATP infusion and exercise is the influence of the muscle pump aiding venous return during exercise. Although cardiac output was matched between conditions, it is unknown if venous return was equivalent between the ATP and exercise conditions, which may differentially influence heart rate and/or SV. In fact, central venous pressure was significantly lower in the ATP infusion trial compared with rest and exercise during the higher atrial pacing levels. In this context, it is remarkable that a local ATP infusion alone can mimic the tissue perfusion and cardiac output that occurs during one-leg knee-extensor exercise. Previous work has demonstrated that ATP infusion or purinergic stimulation overrides tyramine-induced muscle sympathetic vasoconstrictor activity (Rosenmeier et al. 2008), whereas sympathetic activity is increased during exercise. There is evidence that ATP infusion increases muscle sympathetic activity in humans, and both interstitial and arterial ATP stimulate the release of noradrenaline from the nerve terminal in rats. However, in the current study, noradrenaline concentration during ATP infusion was reduced during the second pacing level and was lower than the exercise trial. ATP can prevent neurotransmission prejunctionally at the peripheral terminals of sympathetic nerves, but this mechanism is unlikely to be caused by an intra-arterial infusion of ATP.

Another possibility is that hyperperfusion of the leg may cause feedback inhibition, similar to the reduction in sympathetic activity reported during systemic hyperoxia. This raises the following question: does hyperperfusion of major tissue beds alter sympathetic nerve activity independent of an ATP infusion? In the study by Bada et al., cardiac output and leg blood flow were matched, but leg vascular conductance, systemic vascular conductance, and cerebrovascular conductance were slightly (but not significantly) higher with ATP infusion compared to exercise, likely to be driven by the lower mean arterial pressure. However, the lower noradrenaline during ATP infusion indicates that sympathetic nerve activity is potentially inhibited by ATP-induced hyperperfusion.

It is important to consider the duration of atrial pacing (at rest, during exercise, and with ATP infusion) and the duration of ATP infusion. The dramatic blood volume shifts that arise during these experimental conditions lead to lower SV, central venous pressure and mean arterial pressure. Furthermore, left ventricular contractility index was similar between rest and femoral ATP infusion, suggesting that peripheral vasodilatation alone is not enough to change contractility. Therefore, sustained ATP infusion combined with the blood volume shift will eventually cause cardiovascular collapse. In this context, peripheral vasodilatation may determine cardiac output, but eventually, other compensatory mechanisms will be required to maintain cardiac output.

The cardiac output in the study by Bada et al. was relatively low during ATP infusion and exercise when compared to rest (rest: 8.2 ± 0.7 l min⁻¹; exercise: 12.1 ± 0.9 l min⁻¹; ATP infusion: 11.2 ± 0.4 l min⁻¹). Therefore an intriguing question arises: are the mechanisms which are active during whole body exercise or multiple-limb ATP infusion the same as those that are active during single-limb exercise or ATP infusion? In other words, does peripheral vasodilatation and central venous pressure determine cardiac output when oxygen demand is high, such as during multiple-limb whole body exercise which can increase cardiac output up to 20–35 l min⁻¹? In running dogs, central venous pressure and peripheral vasodilatation regulate cardiac output (Sheriff & Mendoza, 2004). However, due to the greater need for oxygen delivery to skeletal muscle during multiple-limb exercise and the subsequent demands placed on the heart by the peripheral circulation, the regulatory mechanisms may be different. During whole body exercise, oxygen delivery may be inadequate due to cardiac limitations. For example, leg blood flow during leg exercise decreases by ∼10% when arm exercise is added. Additionally, the respiratory muscles are also competing for cardiac output during whole body intense exercise with ∼15% of cardiac output directed to the respiratory muscles. However, in the work by Bada et al. it is unclear what percentage of maximum effort was used and it is doubtful that this intensity increased the need for respiratory muscle blood flow. Due to cardiac limitations during whole body intense exercise, peripheral vasodilatation will likely be insufficient to fully compensate for a high cardiac output.

Another important consideration during multiple-limb exercise is that heart rate is much greater than single-limb exercise. Shortened ventricular filling time may limit end-diastolic volume at high heart rates and restrict SV. Indeed, SV plateaus at ∼50% maximal exercise. Under this premise, and in agreement with Bada et al., heart rate at least partially dictates SV at higher heart rates. Along these lines, preload is vastly different between single and multiple-limb exercise. Additionally, sympathetic activity is greater during multiple-limb exercise versus single-limb exercise. In this context, functional sympatholysis promotes vasodilatation in active muscle, whereas sympathetically-mediated venoconstriction promotes venous return. Thus, greater preload during multiple-limb exercise would increase cardiac contractility. Although controversial, additional muscle groups may also enhance venous return via the muscle pump. In the study by Bada and colleagues, higher heart rates coupled with lower end-diastolic volumes during single-leg exercise caused a decrease in contractility index. The net effect of augmented heart rate and reduced end-diastolic volume during multiple-limb exercise might modulate SV while maintaining cardiac output.

Many studies investigating cardiovascular control during exercise are conducted in animals or young healthy humans. Therefore, additional cardiac regulatory mechanisms may become necessary when a subject has impaired peripheral vasodilatory responses (i.e. peripheral artery disease or diabetes). Individuals with impaired peripheral vasodilatation and cardiac output may not be able to meet metabolic oxygen demand. Instead, these individuals may rely on heart rate to increase cardiac output because vasodilatation cannot fully contribute. Thaning et al. (2010) demonstrated that type 2 diabetics have an attenuated sensitivity to vasodilatory agents despite having similar purinergic receptor density as that of healthy subjects. The same study showed that cardiac output and SV were higher in diabetic subjects compared to healthy controls at rest, indicating that SV may over-compensate for impaired peripheral vascular function. During the infusion of vasodilatory substances and during one-leg knee-extensor exercise, SV remained higher in diabetics and cardiac output tended to be greater, indicating that diabetics (and possibly other individuals with endothelial dysfunction) may function with a greater cardiac output compared to healthy controls, and peripheral vasodilatation is not the primary mechanism regulating cardiac output. While diabetics have impaired vasodilatation, they are still capable of increasing cardiac output with activity; however, impaired peripheral vasodilatation may affect their ability to tolerate higher intensities of exercise due to low leg blood flow.

In summary, the work by Bada et al. (2012) answers important questions regarding the regulation of cardiac output and stimulates additional questions for further research, including: Does peripheral vasodilatation regulate cardiac output during whole body exercise? Can peripheral vasodilatation regulate cardiac output over a prolonged period of exercise? Does attenuated peripheral vascular function influence cardiac output and reduce exercise tolerance? In conclusion, the work by Bada and colleagues presents evidence that the complex regulation of cardiac output is modulated by peripheral vasodilatation during low intensity short duration exercise.