Although the effects of exercise training on vascular function have been well studied, less is known about the effects of acute exercise bouts. This synthesis summarizes and integrates knowledge derived from papers relating acute impacts of exercise on artery function, specifically endothelial function assessed by flow-mediated dilatation (FMD). We propose that an immediate decrease in FMD (“nadir”) occurs soon after exercise cessation and that this is followed by a (supra)normalization response. The magnitude of the nadir and (supra)normalization and duration of this biphasic pattern of response appears to be influenced by numerous factors, including the nature of the exercise stimulus (e.g., type, duration, intensity), the subject population (e.g., trained vs. untrained), and various methodological factors. The impact of these factors on the biphasic pattern are most likely mediated through stimuli that underpin altered FMD postexercise, including shear and oxidative stress, changes in arterial diameter, and antioxidant status. We propose that a combination of these stimuli act synergistically to balance the vasomotor responses postexercise. Finally, we discuss the potential (clinical) relevance of the biphasic response after acute exercise, as the immediate nadir may represent an essential response for subsequent training-induced adaptations but may also represent a transient period of increased cardiovascular risk leading to the “exercise paradox.”
- nitric oxide
The impact of exercise training on vascular function has been extensively studied, and recent reviews for both clinical (2) and healthy populations (28) have highlighted that regular exercise training reduces cardiovascular (CV) risk (60). Improvements in traditional risk factors, such as blood pressure and blood lipids, do not fully explain the exercise-mediated reduction in risk (29). It has therefore been hypothesized that this “risk factor gap” may be associated, at least partially, with a direct effect of exercise training on the arteries that engenders improved “vascular health” (27, 45). This hypothesis fits with previous observations in healthy subjects and diseased groups in whom prolonged exercise training typically improves “vascular function and health,” but does not always alter cardiovascular risk factors (29).
The mechanisms by which exercise training results in improved artery health may be related to improvements in the function and health of the endothelium. The endothelium is a single layer of cells that lines the entire cardiovascular system and produces vaosoactive hormones, including nitric oxide (NO). When released from the endothelium, NO results in smooth muscle cell relaxation and artery vasodilation. In addition to its vasodilator effects, NO is a potent antiatherogenic agent that decreases platelet aggregation and vascular permeability and has antithrombotic properties (for review, see Ref. 39). A key stimulus for the release of NO from the endothelium is increased blood flow or, more specifically, the frictional force along endothelial cell membranes, also known as shear stress. Repeated episodic increases in blood flow and shear stress that occur during bouts of exercise are likely to be a key mechanism for adaption in vascular function and the positive remodeling that results from exercise training (50).
Interestingly, different modes of exercise produce different patterns of arterial shear (79), and differing shear patterns result in distinct responses in vascular function (81). More specifically, increases in antegrade shear are typically associated with improvements in vascular function and upregulation of the NO pathway (81), whereas increases in retrograde and/or turbulent shear may result in decreases in NO bioavailability and a generation of a proatherogenic phenotype (for review, see Ref. 62).
Despite the clear link between exercise training, vascular function, and health, the acute effects of exercise on vascular function have received relatively little attention. Acute exercise presents important challenges to the cardiovascular system, which, when performed repeatedly during exercise training, eventually induce adaptations. This concept is encapsulated by the “hormesis” hypothesis, a physiological concept that holds that upregulation or improvement in physiological parameters (e.g., vascular function) to a repeated stimulus (e.g., exercise training) can be induced if these stimuli challenge and temporarily impair the physiological system. That is, the initial challenging stimulus leads ultimately to activation of beneficial adaptive processes (66).
ACUTE EXERCISE AND ENDOTHELIAL FUNCTION
The majority of studies examining acute exercise and vascular function have assessed flow-mediated dilatation (FMD). FMD reflects conduit artery diameter change (using high-resolution ultrasound) in response to an increase in blood flow (or shear rate). In humans this is usually assessed in the brachial or femoral artery by creating a shear stimulus via reactive hyperemia induced by a period of cuff-induced distal limb ischemia (usually 5 min). This dilator response is likely to be endothelium dependent (16, 17) and thought to be largely NO mediated (40, 48, 61, 78). Compared with plethysmography, ultrasound has high temporal resolution, allows for repeated measurements, and can be undertaken simultaneously in different arteries. This technique is therefore highly suitable for examining vascular function immediately after exercise. This review will focus on studies that have used FMD to assess endothelial function after a bout of exercise in healthy humans. There are clear differences in the findings of these studies that we believe can be explained by inconsistencies in methodological approaches and the types of exercise adopted.
Acute Bouts of Exercise and FMD: Impact of Timing of Postexercise Measurements
Most studies have examined endothelial function at a single time point, varying from immediately after exercise (5, 13, 24, 37, 41, 51, 59, 68, 75, 83, 86, 90), through 0.5–2 h postexercise (18, 23, 33, 35, 43, 46, 64, 72, 73), up to 24–48 h postexercise (69, 89). Not surprisingly, given the differences in methodology, these studies reported conflicting results, ranging from an increase, to a decrease, or no change in FMD.
When examining the studies that have measured FMD at a discreet time point some (5, 18, 22–24, 37, 43, 51, 59, 68, 73, 83, 90), but not all (13, 18, 23, 37, 41, 43, 68, 73, 75, 83, 86, 90), have found a decrease in FMD when it is measured less than 30 min postexercise. These differences are likely to relate to the subject population, the exercise duration, mode, and intensity. The influence of these different methodological approaches on postexercise FMD responses will be discussed in more detail throughout the synthesis. Similarly, measurements taken 30–60 min postexercise are conflicting, with some reporting improvements (22, 33, 35, 42, 94) and others reporting decreases (33, 46, 71) or no change (35, 42, 71) in FMD after exercise. As various other methodological differences, such as exercise intensity, duration, modality, or subject fitness, exist between these “one time point” studies, it is instructive to focus on the few studies that have repeated their measurements after an acute bout of exercise (9, 22, 42, 47, 71, 94) to assess the time course of change in the response.
Studies that have taken multiple FMD measures after exercise suggest that the timing of the postexercise FMD assessment can alter the findings. Such studies (9, 22, 42, 47) typically report an immediate decrease in FMD after the exercise bout, which then normalizes at later time points. For example, Johnson and colleagues (42) found a decrease in FMD immediately after high-intensity moderate-duration running exercise that normalized after 1 h. Our recent data indicated that FMD decreased immediately after 30 min of both moderate-and high-intensity cycle exercise. FMD responses returned to baseline levels by 60 min (9). This further supports the finding of a biphasic change in FMD after exercise. In line with the biphasic response of the postexercise FMD, the majority of studies, but not all (71, 94), reported a postexercise improvement in FMD when examined greater than one h postexercise (13, 22, 33, 42, 64, 68, 72, 83, 86, 89, 94).
Overall, these data suggest the presence of a biphasic response for the postexercise FMD as argued by Gonzales et al. (23), with a decrease in FMD immediately after exercise followed by a normal to supranormal FMD and ultimate normalization within 24–48 h after exercise. Importantly, this biphasic pattern is in broad agreement with animal data that demonstrated, in rats, that a biphasic change in endothelium-dependent responses to acetylcholine is present after exercise (32). It is important to note that several studies, have examined dilation in response to a sublingual dose of glyceryl trinitrate (GNT) pre- and postexercise in addition to FMD responses (51, 68, 71, 75, 86). GTN represents endothelium-independent function and as such reflects the ability of the smooth muscle to dilate to an increase in NO. Although most reported changes in FMD (35, 51, 68, 71, 86), only one reported a change in GTN after acute exercise (35). This later study found an exercise-induced decrease in both FMD and GTN in postmenopausal women after estrogen supplementation, with no change in either measurement with exercise pre-estrogen supplementation (35). Overall, these responses suggest that any alterations reflect a reduction in endothelium-dependent NO function rather than any alteration in smooth muscle function.
In summary, the timing of the postexercise measurement is an important factor that contributes to the current disparity in the literature. Studies examining postexercise responses in FMD should, therefore, preferably include multiple postexercise assessments. In the following sections, we discuss factors that may alter the nature, strength, or direction of this biphasic FMD response (Fig. 1).
Acute Bouts of Exercise and FMD: Impact of Exercise Intensity and Duration (Aerobic Exercise)
The most common form of exercise adopted to assess acute impacts on FMD is aerobic exercise. However, the intensity and duration of this exercise vary markedly, which appears to significantly modify the nature of the biphasic, postexercise change in endothelial function (Fig. 1). Higher exercise intensities (>80% V̇o2max) typically result in a larger decrease in FMD immediately postexercise (9, 42), whereas most (13, 33, 35, 41, 64, 86, 89, 94), but not all (68, 72, 83, 89), studies of low-moderate intensity exercise (50–80% V̇o2max) have reported an increase in FMD after exercise.
In a recent study, Birk et al. (9) directly examined the influence of exercise intensity on postexercise FMD, by administering three exercise bouts at different intensities in each participant. They observed an exercise intensity-dependent response, with immediate decreases in FMD after high, but not low, exercise intensities. Similarly, Johnson et al. (42) reported a decrease in FMD after a high-intensity exercise bout but not after moderate-intensity exercise (42). Mills et al. (59) also found a decrease after high-intensity exercise, but not low intensity, in children. In addition, these studies all examined FMD immediately after exercise and as such were more likely to capture the putative early drop in postexercise FMD biphasic response. It is important to acknowledge that not all studies have reported an impact of exercise intensity on the magnitude of postexercise changes in FMD (33). However, in the study of overweight men by Harris et al. (33), FMD measurements were taken 1 h postexercise, i.e., when normalization of initial changes in FMD may have occurred (33). Interestingly, this latter study also reported an impact of training status on 1 h postexercise FMD responses, with fitter subjects exhibiting enhanced FMD and less fit overweight subjects exhibiting depressed FMD 1 h after exercise. This suggests that, in addition to exercise intensity, physical fitness may also contribute to the pattern of the postexercise biphasic response, with a higher fitness associated with an attenuated decrease in FMD immediately postexercise.
Apart from intensity, factors such as the duration of the exercise bout may also modulate the FMD response. Studies examining postexercise changes in FMD after shorter duration exercise tests have typically used maximal oxygen uptake tests (V̇o2max tests). The data for FMD after this mode of exercise is conflicting with increases (83), decreases (22, 37, 47), and no change (37, 74, 83) in FMD reported. The contradictory results may relate to either the short duration (typically <20 min), the brief period of time spent at high intensity and the uncontrolled stimulus that they produce.
Interestingly, high-intensity exercise of longer duration typically leads to a decrease in FMD (in the initial biphasic response phase) (9, 42), whereas short bursts of (a similar) intense exercise may be of insufficient duration to induce this response (37, 42, 74, 83). In support of this, Johnson et al. (42) did not observe a decrease in FMD immediately after a short and intense exercise bout but found an immediate decrease in the same subjects when they performed exercise of similar intensity for a longer period. Further analysis of the impact of exercise duration is difficult because very few studies have used exercise of greater than 1 h.
Overall, the studies described above suggest that exercise intensity influences the immediate postexercise FMD responses, with higher exercise intensities eliciting greater immediate decreases in FMD in healthy subjects. However, it is important to emphasize that some data suggest that this relation is only present when the duration of the exercise bout is sufficiently long. The influence of exercise intensity on measurements taken longer than 1 h postexercise is less clear. Johnson et al. (42) manipulated both exercise intensity and duration and argued that the improvement in FMD at 1 h postexercise is associated with neither the intensity nor duration, but rather the exercise “dose” estimated on energy expenditure. In contrast, a recent paper by Hwang et al. (37) found no relationship between change in FMD and METs during exercise. Further studies are therefore needed to understand the precise and independent impacts of exercise intensity and duration on FMD during both the first and the second phase of the postexercise response.
Acute Bouts of Exercise and FMD: Impact of Mode of Exercise
In addition to the strong focus of most studies on aerobic exercise, there have been several studies that have examined resistance exercise (23 –24, 46, 68, 68, 90). Resistance exercise is typically accompanied by larger increases in blood pressure (46, 54), and it is possible that any potential decrease in FMD seen after this mode of exercise is associated with the elevated blood pressure that has been shown in vitro to reduce vascular function (7). In support of this, most studies have reported a decreased FMD after resistance exercise (23, 24, 46, 68, 90). However, this may also be due to the fact that the majority of these studies examined FMD within the first 30 min postexercise (23, 24, 68, 90). It is interesting to note that in these same studies decreases in FMD are less apparent if the subjects are trained (46, 68, 86, 90). It is therefore suggested that exercise training protects the artery from the immediate decreases in FMD after acute exercise (37, 46, 68, 90), a putative “conditioning” effect on the vasculature.
POTENTIAL MECHANISMS UNDERLYING POSTEXERCISE CHANGES IN FMD
Oxidative stress is an imbalance that favors pro-oxidant status over antioxidants, which may lead to vascular damage and/or dysfunction. There have been numerous studies that have highlighted a negative relationship between increased oxidative stress and vascular function under baseline conditions (for review, see Ref. 20). It is hypothesized that an increase in oxidative stress can ultimately lead to a reduction in bioavailability and production of nitric oxide, a key factor in FMD responses (63). More specifically, NO scavenges free radicals, reducing NO availability (10). The reaction between superoxide anion, a key free radical produced during exercise, and NO results in the formation of peroxynitrite (6), a potent oxidant that decouples eNOS by oxidizing the zinc-thiolate center (71). Thus exercise increases oxidative stress, which may lead to an even greater reduction in the bioavailability and production of NO and subsequent reduction in endothelial function and FMD (71).
On the basis of the production of oxidative stress during exercise, the reduction in FMD immediately after an acute exercise bout has been attributed to this mechanism (22, 33, 51, 75). Supportive of this hypothesis is that higher exercise intensities are associated with a greater oxidative stress (25) but are also related to an exaggerated decrease in FMD (9). Johnson et al. (41) recently reported improved brachial FMD 1 h after moderate intensity cycling under normal conditions. However, when exercise was accompanied by an increase in retrograde shear, provoked by cuffing around the forearm during exercise, the FMD was attenuated. Interestingly, this attenuation was abolished with an antioxidant supplement (i.e., vitamin C) that typically lowers oxidative stress. This suggests that increased oxidative stress induced through increases in retrograde shear was responsible for the attenuation in postexercise FMD. Further support for this hypothesis comes from Silvestro et al. (75), who reported that a decrease in FMD after exercise was prevented by decreasing oxidative stress through antioxidant supplementation. This finding, however, was limited to patients with peripheral artery disease, as healthy control subjects exhibited no decrease in FMD or effect of antioxidant supplementation. Habitual exercise training improves antioxidant status (for review, see Ref. 20), and studies of acute effects of exercise on FMD suggest that trained individuals, therefore, exhibit a smaller decrease in FMD immediately postexercise. Indeed, the majority (33, 37, 46, 68, 90), but not all, studies (71) reported a decrease in FMD after exercise in untrained subjects, with no changes in FMD apparent in trained individuals.
Not all studies have reported a clear relationship between oxidative stress and FMD after acute exercise. Somewhat in contrast to their earlier study (41), Johnson et al. (42) found no relationship between the postexercise FMD and measures of oxidative stress [as assessed by thiobarbituric acid reactive substances (TBARS)] after a range of exercise intensities and duration. However, an important methodological limitation associated with assessment of oxidative stress in relation to acute exercise relates to the single measurement of oxidative stress (42, 71) rather than the rate of antioxidant depletion (4, 42). In addition, Johnson et al. (42) highlight that the inability to measure oxidative stress at the site where the FMD is measured, and the likelihood that nitric oxide bioavailability is decreased prior to the appearance of markers of oxidative stress may explain the lack of correlation found in some studies. Therefore, development of oxidative stress during exercise likely contributes to an immediate decrease in FMD after a single exercise bout, especially because protection against oxidative stress (through exercise training or supplementation) prevents against this response (41, 46, 68, 75).
The late phase of the postexercise FMD response (Fig. 1) may be related to changes in (anti)oxidative status. The immediate increase in oxidative stress during exercise, potentially associated with initial FMD impairment, may trigger an increase in the production of antioxidant enzymes (32, 38, 89). Tyldum et al. (89) argued that during exercise there is a transfer of antioxidants into the muscle, tipping the balance of oxidative status toward an increased antioxidative state. If this is true, then FMD responses assessed after a delay may increase as a result of a more favorable oxidative milieu. This hypothesis requires further focused research.
Influence of Shear Rate
Mean shear rate.
During exercise there is a large increase in blood flow supplying active muscles in response to an increased metabolic demand (26). This results in an increase in antegrade shear across the endothelium, which stimulates the release of vasoactive substances such as NO (86). In an animal model, Haram et al. (32) reported that, after exposure to a single exercise session, rat aortas had increased calcium influx into endothelial cells, with improved NO vasodilation for 48 h. Improved FMD after exercise may therefore be associated with improved NO bioavailability due to increased shear stress (13, 35, 64, 69, 86, 94).
However, as noted above, decreases in FMD have commonly been reported immediately after exercise. The biosynthesis of NO is dependent upon the continual availability of l-arginine (39) and several cofactors. Typically, arginine stores do not limit NO production in healthy resting humans (70). However, substrate utilization is a known limiting factor in states that involve elevated oxidative stress (63). Although speculative, it is possible that the availability of substrates may be limited during prolonged high levels of exercise-induced shear or oxidative stress (9, 18), potentially leading to a “fatigue” in the capacity to produce NO and an inability to dilate in response to a given shear stimulus. Dawson et al. (18) is one of the few studies to examine FMD after prolonged intense exercise (>1 h). They reported a decrease in FMD in the superficial femoral artery but no change in FMD of the brachial artery after a competitive marathon. An important difference between both arteries is that the superficial femoral artery, by supplying the highly active calf muscles, received a much larger and continuous shear stress stimulus than the brachial artery, which supplies the less active upper limbs. To determine whether the reduction in FMD postexercise is associated with a reduction in substrate availability, future studies should use arginine supplementation to determine whether this preserves the FMD response.
Another potential mechanism for reduced FMD after exercise may be a reduced shear rate during the FMD test itself. It has been demonstrated that shear rate is an important stimulus for the dilation of arteries during the FMD test (78), although other vasoactive factors may be involved that contribute to the dilation during the FMD response [e.g., adenosine, endothelin, progstaglandins, sympathetic nervous system (78)]. As the artery is already dilated immediately after exercise or because the artery is subjected to continuous high flows during exercise, there may be a decrease in cuff-induced postischemic shear stress (24, 71). The reduced FMD after exercise may therefore simply be associated with a lower stimulus for FMD, rather than an inherent decrease in responsiveness of the endothelium to a shear stimulus. However, the eliciting shear during the FMD test is frequently shown to be either unaltered or increased even when FMD is reduced (5, 9, 18, 33, 42–43, 51, 68), which would argue for an inherent decrease in vascular function associated with substrate limitation (secondary to sustained elevations in shear or oxidative stress).
Different modes of exercise result in different shear rate patterns (79). Moreover, recent work in humans has demonstrated that there is a time-dependent change in the shear pattern of the brachial artery during lower limb exercise (76). The first 5–10 min of moderate intensity exercise is associated with an increase in retrograde shear rate, subsequently followed by a normalization of the retrograde shear rate. The normalization of retrograde shear rate coincides with theromoregulatory vasodilation of the forearm skin (76). These time-dependent changes in shear pattern and/or thermoregulation may influence the postexercise FMD response. Increases in antegrade and laminar flow are associated with antiatherogenic effects and normal vasodilator responses (for review, see Ref. 14), whereas oscillatory and retrograde shear leads to a proatherogenic state (34). Tinken et al. (86) reported that the magnitude of increase in antegrade shear rate was associated with improvements in FMD, while Thijssen et al. (81) reported that increasing levels of retrograde shear resulted in a decline in FMD. Johnson et al. (41) also found that increases in retrograde shear during exercise attenuated postexercise increases in FMD compared with a control condition. Data from animal and in vitro work suggests that nitric oxide is significantly reduced during periods of retrograde flow. This has been demonstrated to be mediated by an increase in superoxides, through activation of the NADPH oxidase system, rather than a reduction in NO production (21, 52). However, the majority of studies examining FMD after an acute bout of exercise have not examined blood flow patterns during exercise, making it difficult to determine the exact role of shear rate during exercise on subsequent postexercise FMD.
In summary, vascular shear and the pattern of shear are clearly potent stimuli to adaptations in vascular function, both acutely and chronically (8, 81–82, 86–87). Further studies are needed to examine the impact of changes in shear patterns during the exercise bout on postexercise measurements in vascular function.
The mechanisms responsible for decreased FMD responses after resistance exercise may differ from those after aerobic exercise. Resistance exercise typically leads to larger increases in blood pressure than aerobic exercise (12, 53, 57). Previous studies have demonstrated that experimentally induced acute hypertension results in an immediate reduction in endothelial function (49, 58). It is possible, therefore, that after resistance exercise, FMD is acutely decreased as a direct consequence of a blood pressure elevation. In support of this, Gonzales et al. (23) reported that a higher blood pressure response during exercise was related to a larger decrease in FMD after handgrip exercise in healthy subjects.
A mechanism by which increased blood pressure decreases FMD may be associated with high levels of shear rate. High levels of blood pressure lead to associated increases in shear that, over a prolonged period, may result in decreased release of NO from the endothelium (7). However, in argument against this hypothesis, Gonzales et al. (23) reported that the decrease in FMD after exercise was associated with elevations in blood pressure, but independent of changes in mean shear. Subjects undertook two bouts of handgrip exercise that consisted of either fast or slow contractions. During slow contractions, blood pressure was higher than fast contractions (23). There was no difference in mean shear rate between the two different contraction speeds. Nonetheless, the slow contractions resulted in decreased FMD, whereas the fast contractions did not, suggesting a mean shear-independent effect of high blood pressure on FMD. However, antegrade, retrograde, and oscillatory shear index were lower during slow contractions compared with fast contractions (23) so an influence of shear pattern on the FMD responses postexercise cannot be ruled out.
As previously mentioned, the decrease in FMD immediately after exercise may be less apparent in trained individuals (33, 37, 46, 68, 90). This may particularly be the case if the exercise results in an elevation in blood pressure (such as resistance exercise). Interestingly, this training-associated protection is apparent even if the exercise training stimulus is not resistance exercise, i.e., if the subjects were runners or cross trainers as opposed to weight lifters (68). The impact of exercise training on postexercise FMDs is likely associated with improved NO production, antioxidant capacity, or improved adipokine profiles (46, 68, 90), which are upregulated by training and improve and/or protect the artery against endothelial dysfunction. In summary, increases in blood pressure are likely to lead to acute decreases in FMD, whether these increases in blood pressure are mediated by exercise, or nonexercise stimuli (23, 49, 58). However, this effect of exposure to episodes of hypertension may be mitigated in the training state, which consequently protects the artery against increases in blood pressure and associated elevations in shear and oxidative stress that underlie the postexercise decrease in endothelial function.
Baseline Artery Diameter
The substantial increases in shear associated with acute bouts of exercise lead to conduit artery dilation. Recently, it was demonstrated that exercise-mediated increases in shear have a strong and dose-dependent effect on conduit artery dilation (11, 67, 92, 93). As baseline diameter is included in the FMD calculation, exercise-induced changes in baseline diameter should be taken into account when examining postexercise FMD. Indeed, numerous studies have reported a strong and inverse relationship between baseline artery diameter and FMD, with larger arteries having diminished dilator capacity than smaller arteries (71, 80). Although it was originally argued that this was due to a larger shear in smaller arteries in response to an ischemic challenge, recent data have suggested that smaller arteries are inherently more sensitive (85). This is possibly related to the observation that they have more smooth muscle relative to the elastic elaminae and, therefore, have a larger vascular reactivity (80, 85). However, it is important note that these comparisons were made between arteries of different size and structure, whereas the acute exercise responses relate to acute changes within the same artery.
It has been suggested that the changes in baseline diameter that result from acute exercise may be a limitation for using FMD to evaluate endothelial function immediately postexercise (3, 65). As larger arteries are associated with a lower FMD, the decrease in FMD after exercise may not represent a decrease in vascular function per se, but rather a consequence of a “diminished dilator reserve” (24). In support of this, Gori et al. (24) examined radial artery FMD after isometric handgrip exercise. Although FMD was reduced immediately postexercise, the constriction of the artery during the cuff occlusion [termed low flow-mediated constriction (L-FMC)] was greater postexercise compared with preexercise. This lead them to hypothesize that rather than representing endothelial dysfunction, a reduced FMD may be attributable to the difficulty to further dilate an artery that is already vasodilated or to recruit an already stimulated endothelium (24). Several previous studies report that the decrease in FMD immediately after exercise is associated with a concurrent increase in artery diameter prior to the cuff occlusion (5, 24, 37, 42, 47, 71, 83). However, both Birk et al. (9) and Katayama et al. (47) recently demonstrated that a decrease in FMD immediately postexercise was still apparent when changes in artery diameter were statistically accounted for. Similarly, several studies reported a decrease in FMD in the absence of change in resting artery diameter after an exercise bout (23, 43, 46, 68). Likewise, some reported improvement or no change in FMD with an increase in baseline diameter (18, 71, 89, 94), suggesting that altered baseline diameter is not the sole reason for changes in FMD after acute exercise. Therefore, baseline diameter typically increases immediately in response to exercise, but these changes in diameter do not fully explain postexercise changes in FMD.
A further difficulty with assessing FMD after exercise is the likely increase in sympathetic nervous activity (SNA), particularly if exercise intensity is moderate to intense (91). This is pertinent because an increase in sympathetic activity is associated with a decrease in FMD (19, 36). The mechanisms by which an increase in SNS may reduce FMD are not fully understood but may be related to a change in baseline diameter, a decrease in the availability of NO, or a decrease in shear stimulus (36). However, the impact of SNS on postexercise FMD responses has not been well studied, particularly as it is difficult to directly measure sympathetic nervous activity during and after exercise (e.g., microneurography assessment of muscle sympathetic nervous activity). It is likely that postexercise responses are transient and variable, with even a decrease in SNA being reported after exercise (31). It is therefore possible that these changes in SNS influence and contribute to the variable response in FMD immediately postexercise and that the timing of the postexercise measurement and the exercise intensity may alter the influence of SNS on postexercise FMD. However, to the best of our knowledge, no study has examined the impact of sympathetic nervous activity (directly measured) on postexercise FMD. Further work is therefore needed to fully elucidate this relationship.
Although the mechanisms mentioned above are the most commonly discussed and examined, other mechanisms have been suggested in the literature to explain acute impacts of exercise on FMD. Other potential mechanisms that relate to postexercise changes in FMD include inflammation (64), changes in vasoconstrictors (23, 42), smooth muscle dysfunction (23), anti-inflammatory cytokines (64), environmental factors such as pollution (73), differences in responses between arteries (18), and estrogen level (35), and these possibilities warrant more detailed investigation.
Summary of Mechanisms
Various mechanisms have been suggested to explain changes observed in FMD after exercise. On the basis of current evidence, we propose that, rather than just one mechanism, there are numerous factors that can alter the biphasic response pattern in FMD after exercise (Fig. 1). We hypothesize that a balance is present between mechanisms that positively or negatively influence vascular function (Fig. 1). This balance is dependent on the characteristics of the exercise bout (mode, duration and intensity) as well as the subject population (e.g., healthy vs. disease or trained vs. untrained).
FUTURE DIRECTIONS AND CONCLUSION
It is important that future studies that perform FMD measurements before and after exercise should adhere to guidelines for performance and analysis of FMD (78). An excellent review has discussed the limitations of such methodological aspects with regards to the acute exercise model in detail (65). An important gap in our current understanding relates to the characteristics of the exercise bout because these characteristics impact upon the change in vascular function after the exercise bout. On the basis of current evidence, we propose that vascular function after an acute exercise bout follows a biphasic pattern, with exercise acutely leading to a transient decrease in vascular function. Further research is necessary to understand how different patterns of postexercise functional responses translate into optimal training-induced adaptations. Most previous studies have examined the brachial artery FMD in response to lower limb exercise. We previously demonstrated that differing flow/shear patterns occur during different modes of exercise (79). Therefore vessels in the active vs. the nonactive area may respond differently after an acute exercise bout. However, to the best of our knowledge, there are only two studies that have examined FMD in the lower limbs after an acute exercise bout (18, 84). Both reported FMD at one discreet time point and reported distinct findings in the change in FMD (18, 84). Therefore, given that regional differences in upper and lower limbs are apparent with no correlation between FMD in the different vascular beds (84), it would be useful to determine if a postexercise biphasic response also exists in the lower limbs. Another focus for future studies is the clinical relevance of the immediate changes in vascular function after exercise for ultimate vascular adaptations when the exercise bout is repeated during exercise training. The immediate decrease in FMD after exercise may represent the initiation of an adaptive response (30, 66, 77). This hypothesis is in agreement with many examples in integrative human physiology, where upregulation in response to stimuli occur after (repetitive) exposure to an acute challenge, a notion encapsulated in the concept of hormesis. The observation that exercise training itself can attenuate the immediate postexercise decrease in FMD provides further support for this idea (33, 37, 46, 68, 90). Alternatively, it is possible that reductions in vascular function immediately after exercise may represent an “exercise-induced vascular fatigue” (18) that does not eventually lead to an improvement in vascular function. Finally, the acute decrease in FMD after exercise may relate to (a transient period of) increased cardiovascular risk, particularly in susceptible individuals (42). Although exercise training is beneficial to CV morbidity and mortality, there is an increased risk of an acute cardiac event during exercise, in accordance with “exercise paradox” (55). Acute changes in blood flow, shear rate, and oxidative stress may contribute a transient period of elevated vascular risk (1). Although this review has focused on healthy individuals, there have been a number of studies that have examined the impact of an acute exercise bout on FMD in clinical populations. As with the data in healthy individuals, the responses are likely to vary with exercise intensity, duration, and timing of measurements postexercise. This relationship is further complicated by the different clinical populations examined, the degree of their disease progression and the interactive effect of therapeutic drugs on vascular function. Nonetheless, a number of studies demonstrated a decrease in FMD immediately after exercise (30, 44, 56, 75), although there were some who reported an increase (15, 88). A biphasic response has been reported with a decrease at 30 min, which is still depressed at 2 h but recovered by 4 h in patients with peripheral artery disease (PAD) (44). However, in contrast to this, Tjonna et al. (88) found that FMD was increased immediately postexercise and remained elevated for 24–72 h later. Further studies employing multiple measurements postexercise in a range of patient populations are therefore needed to determine whether a biphasic response exists and to what extent the disease process influences it. There is also evidence that exercise intensity, similarly to healthy subjects, influences the postexercise FMD response as Silvestro et al. (75) reported an immediate decrease in FMD with intense but not moderate exercise in patients with PAD.
A possible mechanism for decreased FMD immediately after exercise, particularly in patients with PAD, who are likely to have high levels of ischemia during exercise (30, 44, 75), is the increased oxidative stress leading to an increased degradation or reduced synthesis of NO. In support of this, Silvestro et al. (75) reported that the decrease in FMD was prevented with vitamin C supplementation in patients with PAD. Likewise, Gresele et al. (30) found, in a similar patient population, that the immediate decrease in FMD after exercise is reduced with nitroaspirin supplementation. Conversely, the improved FMD seen at later time points with exercise in patient populations may relate to the initial lower eNOS levels at baseline and subsequent low FMD pre-exercise. Subsequently, the exercise-induced increase in shear is likely to lead to an increase in NO production and potential improvement in FMD. Future studies are necessary to better understand the clinical and physiological relevance of the postexercise changes in FMD.
The acute effects of exercise provide insight into the causes of longer-term adaptation. At present, the relevance of acute changes in vascular function to long-term adaptations is not known. Large variations in methodology are present between the studies we reviewed. Nonetheless, the current literature suggests that the timing of the measurement influences the findings and we hypothesize the presence of a biphasic response in FMD after exercise (Fig. 1). This response pattern is broadly in keeping with known changes in postexercise oxidative stress and inflammation and with animal studies on the time course of changes in vascular function after exercise (32). The nature of the vascular function response to exercise seems to depend on the exercise stimulus (type, duration, intensity), subject group (trained vs. nontrained, healthy vs. disease; Fig. 1) and methodological factors (diameter changes immediately after exercise). It is likely that a combination of these influences act synergistically to balance the vasomotor response postexercise. Further studies are required to fully understand the nature of the immediate changes in vascular function, as assessed by FMD and other responses, immediately after exercise.
Professor Green is funded by the Australian Research Council (DP 130103793). Dr. Thijssen is financially supported by the Netherlands Heart Foundation (E Dekker-stipend, 2009T064).
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: E.A.D., D.J.G., and D.H.T. conception and design of research; E.A.D., D.J.G., N.T.C., and D.H.T. interpreted results of experiments; E.A.D. prepared figures; E.A.D. drafted manuscript; E.A.D., D.J.G., N.T.C., and D.H.T. edited and revised manuscript; E.A.D., D.J.G., N.T.C., and D.H.T. approved final version of manuscript.
- Copyright © 2013 the American Physiological Society