Much attention has been directed toward lifestyle modifications as effective means of reducing cardiovascular disease risk. In particular, physical activity has been heavily studied because of its well-known effects on metabolic syndrome, insulin sensitivity, cardiovascular disease risk, and all-cause mortality. However, data regarding the effects of exercise on various stages of the atherosclerosis pathway remain conflicting. The investigators review previously published reports for recent observational and interventional trials investigating the effects of physical activity on markers of (or causal factors for) atherosclerotic burden and vascular disease, including serum lipoproteins, systemic inflammation, thrombosis, coronary artery calcium, and carotid intima-media thickness. In conclusion, the data show a correlation between physical activity and triglyceride reduction, apolipoprotein B reduction, high-density lipoprotein increase, change in low-density lipoprotein particle size, increase in tissue plasminogen activator activity, and decrease in coronary artery calcium. Further research is needed to elucidate the effect of physical activity on inflammatory markers and intima-media thickness.
Despite the strong evidence linking physical activity to cardiovascular disease (CVD) risk reduction, there remains much uncertainty regarding the mechanisms underlying the effect of exercise on progression of atherosclerosis. In this review, we summarize recent data on the effects of physical activity on causal factors for atherosclerosis, including atherogenic lipoproteins, systemic inflammation, and thrombosis. We also review recent studies on the effects of exercise on downstream markers of atherosclerotic burden, including vascular calcium and carotid intima-media thickness (IMT).
Effect of Physical Activity on Causal Factors for Atherosclerosis
Lipoproteins
The association between serum cholesterol and CVD outcomes is well documented in the published research. In particular, low-density lipoprotein (LDL) and apolipoprotein B have been correlated with the development of coronary artery disease (CAD) and CVD-related events. Early in atherogenesis, proteoglycans within arterial walls are thought to bind LDL, which subsequently results in local inflammation and the initiation of a complex pathway toward atherogenesis. Therefore, reductions in serum levels of atherogenic lipoproteins would be expected to minimize atherosclerotic progression and subsequent CVD risk. Despite the well-established benefits of regular exercise in CVD prevention, and the well-established link between lipids and CVD, there are relatively few data regarding the effect of regular exercise on atherogenic lipoproteins.
To date, most studies noting reduction in total cholesterol or LDL were in the setting of significant exercise-induced weight loss. In fact, most data regarding exercise training and serum lipids seem to indicate that regular exercise training does not significantly reduce total cholesterol or LDL independent of weight loss ( Table 1 ). Randomized trials studying the effect of aerobic exercise and comprehensive rehabilitation in patients with CAD noted no change in LDL or total cholesterol after ≥6 months of intervention. However, data do suggest that regular physical activity may change LDL particle size, even when total LDL concentrations remain constant. Two cross-sectional studies noted lower concentrations of small, dense, atherogenic LDL in patients reporting higher levels of physical activity. LDL density was found to range from 1.040 to 1.063 g/ml in sedentary subjects and from only 1.019 to 1.037 g/ml in more active subjects. Furthermore, in a longitudinal study, Kraus et al found that 25 minutes of daily aerobic activity increased mean LDL particle size (p <0.03) irrespective of training intensity or weight loss. Given the atherogenic potential of small, dense LDL particles, exercise training appears to reduce CAD risk, in part, because of increases in LDL particle size rather than significantly lowering total LDL concentrations.
| Study | Design | Sample Size | Intervention Measure | Significant Outcomes |
|---|---|---|---|---|
| Kokkinos et al | Cross-sectional | 2,906 men | Self-reported equivalent of miles jogged per week | Positive correlations between miles run per week and HDL-C and LDL-C |
| Thompson et al | Interventional | 17 men | Supervised training: 1 hour aerobic exercise 4 times/week for 1 year | HDL increased by 10%, Apo A-I increased by 9%, TG decreased by 7%, Apo B decreased by 10% |
| Belardinelli et al | Randomized | 118 CAD patients | Exercise 3 times/week for 6 months vs control | Decrease in total cholesterol, LDL, and TG in intervention group; increase in HDL cholesterol in intervention group |
| Yu et al | Randomized | 112 CAD patients | Comprehensive cardiac rehabilitation program for 2 years vs control | Decrease in TG in interventional group; no change in total or LDL cholesterol |
| Wosornu et al | Randomized | 81 men after heart surgery | No exercise vs 6 months aerobic exercise vs 6 months strength training | No changes in lipoprotein levels with interventions |
| Halle et al | Cross-sectional | 40 men | Self-reported exercise >3 times/week vs sedentary | Lower TG and higher HDL in trained subjects; fewer small, dense LDL particles in trained subjects |
| Kraus et al | Randomized | 111 adults | Control vs 3 exercise categories grouped by exercise intensity and amount | Decrease in TG, increase in HDL, and decrease in number of small, dense LDL particles in exercise groups |
| Ziogas et al | Cross-sectional | 54 adults | Self-reported activity | Lower TG, lower Apo B100 in trained subjects |
| Mora et al | Cross-sectional | 27,158 women | Self-reported activity | Lower total cholesterol, LDL, Apo B, and higher HDL in active subjects |
| Alam et al | Randomized | 18 diabetics | Supervised vs unsupervised exercise 4 times/week at 70% peak V o 2 | Lower TG and Apo B in supervised exercise group |
| Holme et al | Randomized | 188 men | Diet vs endurance training 3 times/week vs diet and training vs control | Decrease in Apo B with physical activity; no change in LDL |
| Ring-Dimitriou et al | Randomized | 30 adults | Aerobic training for 9 mo vs control | Decrease in Apo B in exercise group |
| Ballantyne et al | Randomized | 42 men after MI | Incremental exercise program vs control | Decreases in LDL and TG, increases in HDL and apolipoprotein A-1 in exercise group |
| Leon et al | Interventional | 675 adults | 20 weeks supervised cycle-ergometer exercise | Decrease in TG, increase in HDL after exercise; no change in LDL, Apo B, or total cholesterol |
| Thomas et al | Interventional | 17 women | Moderate aerobic exercise for 6 months | Increase in HDL; no change in total cholesterol, LDL, or TG |
Results on the effects of exercise on apolipoprotein B, high-density lipoprotein (HDL), and triglycerides are more promising. Several cross-sectional studies have noted lower apolipoprotein B levels in subjects who reported higher levels of physical activity. Longitudinal studies have shown regular exercise to reduce apolipoprotein B by up to 20%. As for HDL and triglycerides, 1 meta-analysis showed that long-term, moderate-intensity exercise training increases HDL and lowers triglycerides even in the absence of weight loss. With 30 to 60 minutes of moderate-intensity aerobic exercise 3 to 5 times per week, HDL levels were noted to increase by 0.05 mmol/L, and triglyceride levels decreased by 0.21 mmol/L (p <0.01). Others have shown that HDL increases by 0.008 mmol/L per mile of running per week. Several other studies have noted similar effects of exercise on HDL and triglycerides, irrespective of changes in body weight.
In light of the strong evidence for changes in apolipoprotein B, HDL, and triglycerides with regular exercise, consensus panels continue to highly recommend exercise training for CVD prevention. Although recent data indicate no change in serum LDL concentrations, there is mounting evidence of alteration in LDL particle size with exercise and therefore a presumed decrease in atherogenicity. Further conclusions regarding ideal exercise duration, intensity, frequency, and dose response of serum lipoprotein and apolipoprotein changes have yet to be established.
Systemic inflammation
Given the central role of inflammation in the pathology of atherosclerosis, inflammatory markers are being increasingly measured for CVD risk stratification. A few of the most commonly investigated markers have been C-reactive protein (CRP), various interleukins (ILs), leukocytes, and fibrinogen. On a molecular basis, exercise and skeletal muscle contraction are thought to acutely increase circulating cytokines including IL-6. IL-6 is subsequently able to inhibit tumor necrosis factor–α, thereby decreasing insulin resistance and stimulating lipolysis. With regular exercise training, basal IL-6 levels are thought to decrease over the long term because of lower glycogen content, improved antioxidative capacity, and decreased insulin resistance. Accordingly, this is expected to result in lower CRP levels with long-term training as well.
Despite abundant evidence behind the aforementioned molecular hypothesis, clinical results regarding the effects of physical activity on various inflammatory markers have been largely inconsistent and conflicting. Although numerous cross-sectional studies have observed an inverse relation between physical activity and CRP levels, only a few have noted a relation independent of obesity or weight change. In prospective studies, regular physical activity was found to decrease CRP by approximately 25% to 50%. Results regarding effects on CRP from randomized controlled trials are more mixed ( Table 2 ). Two randomized controlled trials investigating the effect of 45 minutes of aerobic exercise 3 to 5 times per week for >6 months showed a statistically significant reduction in CRP in healthy subjects compared to controls. Similar results were shown in randomized controlled trials investigating diabetic, CAD, and peripheral vascular disease patients. In contrast, ≥8 randomized controlled trials in the past decade have observed no change in CRP after >6 months of exercise training composed of aerobic exercise and/or weight training. The largest of the exercise intervention studies, the Health, Risk Factors, Exercise Training and Genetics (HERITAGE) Family Study, observed no statistically significant reduction in CRP across 490 participants with baseline CRP levels <3.1 mg/L. Results of randomized controlled trials conducted in the CAD patient population have also been mixed, with some trials showing exercise-induced reductions in CRP and others showing no change.
| Study | Sample Size (Intervention/Control) | Duration | Intervention | Significant Outcomes |
|---|---|---|---|---|
| Kohut et al | 48/49 | 10 months | 45 minutes aerobic exercise 3 times/week vs flexibility/strength exercise 3 times/week | Decreased CRP and IL-6 in both groups |
| Campbell et al | 100/102 | 1 year | 1 hour aerobic exercise 6 times/week vs control | No change in CRP |
| Kadoglou et al | 30/30 | 6 months | 30–45 minutes aerobic exercise 4 times/week vs control | Decrease in CRP, increase in IL-10, and decrease in IL-18 in interventional group |
| Walther et al | 51/50 | 2 years | 20 minutes daily aerobic exercise vs PCI in CAD patients | Decreased CRP and IL-6 in both groups |
| Nawaz et al | 26/26 | 6 weeks | Lower- and upper-body ergometer twice per week vs control in patients with intermittent claudication | No change in von Willebrand factor, E-selectin, CD11b, or CD66b |
| Hammett et al | 30/31 | 6 months | 45 minutes exercise 4 times/week vs control | No change in CRP |
| Nicklas et al | 67/70 | 18 months | 15 minutes weight training and 30 minutes walking 3 times/week vs control in chronic arthritis patients | No change in CRP, IL-6, or TNF-α |
| Marcell et al | 37/14 | 16 weeks | 30 minutes exercise 5 times/week vs control in insulin-resistant subjects | No change in CRP or adiponectin |
| Fairey et al | 24/28 | 15 weeks | 15–35 minutes aerobic exercise 3 times per week vs control in breast cancer survivors | No change in CRP |
| Nicklas et al | 183/186 | 12 months | 150 minutes walking/week vs control | No change in CRP; decrease in IL-6 in interventional group |
| Church et al | 80/82 | 4 months | 150–210 minutes moderate-intensity activity/week vs control | No change in CRP |
| Sixt et al | 13/10 | 1 months | 30 minutes/day submaximal ergometer exercise vs control in patients with impaired glucose tolerance and CAD | No change in CRP |
| Huffman et al | 73/103/86/72 | 6 months | 3 training categories grouped by exercise intensity and amount vs control | No change in CRP |
In observational studies, inverse relations between physical activity and other inflammatory markers, including IL-6, fibrinogen, and leukocytes is again seen. In randomized controlled trials, 3 studies showed decreases in IL-6 with aerobic exercise. One large prospective study found functional fitness and muscle strength to be inversely associated with IL-6 and CRP. Two other studies have noted similar findings of inverse correlation between muscle strength and IL-6 as well as CRP. In contrast, ≥2 large randomized controlled trials have shown no significant changes in IL-6 after exercise intervention. Discrepancies in trial findings were commonly attributed to differences in patient populations, physical activity quantification, inaccuracies of self-reporting, and discrepancies in cardiorespiratory fitness stratification.
In total, there remains strong molecular and observational evidence of an inverse relation between physical activity and markers of systemic inflammation. Studies that have shown this relation have noted greatest reduction in CRP with higher intensity training, particularly in subjects with high baseline CRP levels (>3.0 mg/L). The effects of exercise on systemic inflammation have been less established in randomized trials, with results frequently conflicting in healthy and CAD patient populations. Further research is needed to clarify whether an exercise-induced effect on systemic inflammation exists independent of weight loss, baseline CRP, co-morbid conditions, age, or gender.
Thrombosis
As discussed previously, the presence of excess lipoprotein, vascular damage, and increased systemic inflammation all greatly increase the risk for atherogenesis. An additional critical factor in end-stage progression of atherosclerosis is thrombosis, or vascular clotting. When hypercoagulable factors overwhelm fibrinolytic systems within an atherosclerotic blood vessel, platelets aggregate and initiate clot formation. Acutely, physical activity has been found to simultaneously increase coagulation and fibrinolysis, creating a transiently prothrombotic state. Acute activity also increases platelet aggregation, thereby explaining increased risk for CVD events during physical activity and in the immediate recovery phase.
Despite the acute prothrombotic effects of exercise, several investigators have presented evidence of antithrombotic effects with regular exercise training over the long-term. Two large cross-sectional studies have observed antithrombotic effects with long-term training, particularly due to high circulating levels of tissue plasminogen activator activity. At least 3 prospective studies have also confirmed this increased antithrombotic effect, noting higher circulating tissue plasminogen activator and lower plasminogen activator inhibitor–1 after long-term exercise training. Furthermore, Wang et al have published 2 studies showing decreased platelet aggregation and adhesiveness after long-term, moderate aerobic exercise in male and female subjects. Interestingly, the 2 studies also showed reversal of these thrombotic effects to previous values within 4 weeks of the cessation of exercise.
In summary, transient prothrombotic effects with acute exercise are well documented in the published research, but are thought to reverse with long-term exercise because of increased tissue plasminogen activator activity and decreased platelet aggregation over the long-term. Effects of physical activity on thrombosis are not sustained with cessation of training, and regular exercise must be maintained for patients to continue to benefit from thrombotic risk reduction.
Effect of Physical Activity on Measures of Atherosclerotic Burden
Calcification
Coronary artery calcium (CAC) is an atherosclerotic marker that is strongly linked to risk of future CVD events in adult men and women. Over the past decade, numerous trials have identified associations between CAC and each of physical activity, exercise intensity, and cardiorespiratory fitness. Bishop et al found self-reported physical activity to be independently, inversely associated with CAC in type 1 diabetics. Moreover, several investigators have noted a graded correlation between exercise quantity or intensity and CAC reduction. Storti et al evaluated CAC in 121 postmenopausal women enrolled in a training program and found a statistically significant inverse association between CAC and number of pedometer steps (p = 0.002). Hamer et al found faster exercise walking speed to be associated with lower risk for CAC (odds ratio 0.62, 95% confidence interval 0.40 to 0.96). Similarly, data from the Women on the Move Through Activity and Nutrition (WOMAN) trial and the Multi-Ethnic Study of Atherosclerosis (MESA) showed that walking pace was also associated with lower CAC prevalence in men and women (p <0.05).
Data on cardiorespiratory fitness and CAC are more sparse and conflicting. A prospective trial of 2,373 African American and white young adults from the Coronary Artery Risk Development in Young Adults (CARDIA) study showed that cardiorespiratory fitness was inversely associated with CAC prevalence in young adults (p = 0.03). However, ≥2 well-designed studies have noted no association between CAC and cardiorespiratory fitness after adjusting for gender and other variables. Previous investigators have raised the concern that use of cardiorespiratory fitness as a measure of exercise may not necessarily reflect habitual activity in the long term, because fitness may be influenced by other factors such as co-morbid conditions, genetic background, and functional changes of the vasculature. To our knowledge, no studies to date have quantified exercise duration and intensity and cardiorespiratory fitness simultaneously when investigating their correlation with CAC.
On the basis of these data, there seems to be a strong correlation between exercise quantity and intensity and CAC. Further research is needed to determine whether measures of cardiorespiratory fitness are also independently associated with CAC.
IMT
As atherosclerosis progresses within the vascular wall, the intima-media layers of the arterial wall begin to stiffen and thicken. One large study of 5,858 subjects showed a direct relation between IMT and risk for myocardial infarction (p <0.001). Another recent trial from the Framingham Offspring Study showed that internal carotid IMT >1.5 mm is an independent predictor of CVD events (p = 0.02). Considering the well-known effect of exercise on CVD risk, several investigators have studied the effect of physical activity on carotid IMT. Unfortunately, the results on IMT progression after exercise have been very mixed, particularly in healthy subjects.
Previous observational data showed an independent correlation between cardiorespiratory fitness (as measured by maximum oxygen consumption) and carotid IMT in men and women ( Table 3 ). As for exercise quantification, 2 large cross-sectional studies showed an inverse relation between physical activity and carotid IMT progression in healthy subjects. In the Los Angeles Atherosclerosis Study (n = 500), carotid IMT progression was found to be 14.3 μm/year in sedentary participants, 10.2 μm/year in moderately active participants, and 5.5 μm/year in vigorously active participants. In contrast, ≥3 observational studies have not been able to show a significant association between physical activity and IMT. One study noted an association only for male subjects and attributed this discrepancy to differences in quantification of varying activities between the genders.
| Study | Design | Sample Size | Intervention Measure | Significant Outcomes |
|---|---|---|---|---|
| Tanaka et al | Cross-sectional | 137 adults | Self-reported exercise frequency | No difference in IMT between sedentary and endurance-trained participants |
| Hägg et al | Cross-sectional | 29 adults | Self-reported activity and maximal V o 2 measurement | Inverse association between maximal V o 2 and IMT |
| Moreau et al | Cross-sectional | 77 women | Self-reported endurance training | Lower femoral IMT in endurance-trained women |
| Rauramaa et al | Cross-sectional | 163 men | Maximal VO 2 measurement | Inverse association between maximal V o 2 and IMT |
| Nordstrom et al | Observational | 500 adults | Self-reported leisure-time activity | Decrease in IMT progression from 14.3 μm/year in sedentary subjects to 10.2 μm/year in moderately active and 5.5 μm/year in vigorously active subjects |
| Luedemann et al | Cross-sectional | 1,632 adults | Self-reported diet and physical activity | Lower IMT in subjects with optimal physical activity and dietary patterns and no histories of smoking |
| Schmidt-Trucksass et al | Cross-sectional | 51 men | Self-reported recreational activity | No association between recreational activity and IMT |
| Ebrahim et al | Cross-sectional | 800 adults | Self-reported sporting activity | No association between IMT among groups reporting various activity frequency |
| Folsom et al | Cross-sectional | 14,430 adults | Self-reported exercise and leisure-time activity | Inverse association between workplace activity and IMT |
| Stensland-Bugge et al | Observational | 3,128 adults | Self-reported leisure-time activity | Inverse association between leisure-time activity and IMT only in men |
| Meyer et al | Interventional | 102 children | Intervention: supervised exercise program 3 times/week for 6 months | Decrease in progression of maximum IMT from 0.5% in control group to –8.4% in intervention group |
| Kim et al | Interventional | 58 diabetics | Intervention: 3–12 hours fast walking, hiking, or swimming/week for 6 months | Mean IMT increased by 0.083 mm in control group and showed no change (−0.040 mm) in intervention group |
| Wildman et al | Interventional | 353 women | Intervention: decrease in dietary fat and caloric intake to 1,300 kcal/day, increase of activity to 1,000–1,500 kcal expenditure/week | Decreased IMT progression from 0.008 mm/year in control group to 0.004 mm/year in interventional group |
| Anderssen et al | Interventional | 568 men | Intervention: supervised physical activity and diet for 4 years | No effect of physical activity on IMT progression |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
