This review describes the current state of knowledge on the hazards of exercise and the potential benefits of physical activity on the gastrointestinal tract. In particular, acute strenuous exercise may provoke gastrointestinal symptoms such as heartburn or diarrhoea. A substantial part (20–50%) of endurance athletes are hampered by these symptoms which may deter them from participation in training and competitive events. Nevertheless, these acute symptoms are transient and do not hamper the athlete's health in the long term. The only exception is repeated gastrointestinal bleeding during training and competition, which in the long term may occasionally lead to iron deficiency and anaemia. In contrast, repetitive exercise periods at a relatively low intensity may have protective effects on the gastrointestinal tract. There is strong evidence that physical activity reduces the risk of colon cancer by up to 50%. Less convincing evidence exists for cholelithiasis and constipation. Physical activity may reduce the risk of diverticulosis, gastrointestinal haemorrhage, and inflammatory bowel disease although this cannot be substantiated firmly. Up to now, underlying mechanisms are poorly understood although decreased gastrointestinal blood flow, neuro-immuno-endocrine alterations, increased gastrointestinal motility, and mechanical bouncing during exercise are postulated. Future research on exercise associated digestive processes should give more insight into the relationship between physical activity and the function of the gastrointestinal tract.
Abbreviations used in this paper
- relative risk
- Crohn's disease
- ulcerative colitis
- maximal oxygen consumption
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
The impact of exercise and physical activity on the gastrointestinal tract is an area of emerging interest. During the past two decades research was mainly directed towards the hazards of strenuous exercise, especially gastrointestinal symptoms.1-3In recent years, however, interest has also focussed on the potential benefits of physical activity on the gastrointestinal tract. Several studies indicate an inverse relationship between physical activity and risk of gastrointestinal related diseases such as colon cancer,4-6diverticular disease,7cholelithiasis,8 ,9 or constipation.10 While the prevalence of these diseases is relatively high and increases with age, participation in physical activity is relatively low and decreases with age.
This review summarises the current state of knowledge on the hazards of exercise and the potential benefits of physical activity. “Exercise” is considered as voluntary activation of skeletal muscle leading to short term effects (for minutes or hours) while “physical activity” is considered as repetitive exercise periods leading to long term effects (for days, weeks, months, or years). We focus on the role of physical activity in the prevention of several diseases of the gastrointestinal tract and the postulated mechanisms by which physical activity influences the gastrointestinal tract. These mechanisms will be discussed more extensively at the end of this review. As potential hazards of exercise have been the subject of several reviews (for example, see Moses,11 Peters et al, 12 and Brouns and Beckers13), they will be discussed only briefly.
The role of physical activity in the treatment of gastrointestinal diseases will not be discussed, as the literature on this topic is scarce. Nevertheless, in patients with gastrointestinal diseases physical activity may inhibit muscle loss, and improve appetite, functional capacity, and general well being by positive mood changes5.
Gastrointestinal symptoms such as nausea, heartburn, diarrhoea, and gastrointestinal bleeding are common during exercise, especially during vigorous sports such as long distance running and triathlons.2 ,13-16In general, these symptoms are transient and can be considered protective for critical organ damage: its progressive nature causes the athlete to reduce exercise intensity or duration. Sometimes the symptoms may be so serious that they can limit exercise performance severely1 ,12 ,16and even participation in physical activity.12
Incidence rates during prolonged exercise vary mostly from 20% to 50%, depending on factors such as mode, duration, and intensity of exercise, type of symptom, age, training status, sex, dietary intake, occurrence of gastrointestinal symptoms at rest, and method of investigation.12 ,13 ,15In particular, exercise intensity seems to be an important factor provoking gastrointestinal symptoms.12 ,13 ,16The mechanisms by which exercise causes gastrointestinal symptoms are not well known. Decreased gastrointestinal blood flow, increased gastrointestinal motility, increased mechanical bouncing, and altered neuroendocrine modulation are postulated.11-13All of these mechanisms are associated with exercise intensity.13 ,17-20
While most gastrointestinal symptoms do not hamper the athlete's health, gastrointestinal bleeding may be a serious problem. Most often the type of bleeding is occult and transient, although anecdotal case reports document acute massive upper and lower gastrointestinal bleeding.21 Repeated gastrointestinal bleeding during training and competition may contribute to iron deficiency and anaemia.22 Also, endotoxaemia, malabsorption, gastrointestinal tract inflammation, and hypersensitivity reactions have been postulated to occur.13 ,14 ,21 ,23All of these effects, however, can mostly be prevented by appropriate dietary (for example, sufficient fluid intake) and other precautions (for example, avoidance of large amounts of aspirin and non-steroidal anti-inflammatory drugs).12 ,22
In addition to gastrointestinal symptoms, unfavourable effects of exercise on liver function11 and peptic ulcer disease24 have been reported. However, in well trained endurance-type sportsmen the increase in liver size can be interpreted as a physiological adaptation to increased energy expenditure rather than an expression of liver damage.25 Abnormal high serum levels of bilirubin, aspartate aminotransferase, and alkaline phosphatase are seldom observed in blood, underlining the fact that prolonged strenuous exercise does not lead to serious hepatic damage. Only under extreme exercise conditions, such as heat shock, may hepatic damage occur.25 In the elderly, long term physical activity even may improve liver function.26
Epidemiological studies24 ,27 show that physically demanding occupations may enhance the risk of peptic ulcer disease, independent of several other risk factors such as social class.27 However, these studies were hampered by some limitations and pitfalls which question their conclusions. Recently,28 no relationship between a history of peptic ulcer and leisure time physical activity was found. Physical activity may even contribute to healing of gastric and duodenal ulcers by normalisation of the microcirculation at the ulcer regions.29
The potential benefits of physical activity concern mainly effects on cancer risk, cholelithiasis, gastrointestinal haemorrhage, inflammatory bowel disease, diverticular disease, and constipation.
To date, the risk of oesophageal, bile duct, or gall bladder cancers have not been examined in relation to physical activity. Concerning stomach cancer, the data are controversial: one study reported a reduced risk while two did not.30 No relationship between physical activity and risk of pancreatic cancers has been found,30 ,31 whereas rectal cancer risk was unrelated to physical activity in the majority of studies.4 ,31
In contrast, there is overwhelming evidence that physical activity reduces the risk of colon cancer (for example, see World Cancer Research Fund/American Institute for Cancer Research,4Oliveria and Christos,5 and Colditz and colleagues6). Despite different methods of assessing the amount and different types of physical activity (at work or during leisure time), there is consistent evidence that physically active men and women are at a reduced risk of colon cancer (up to 50% reduction in incidence).6 This effect is independent of other risk factors such as diet and body weight characteristics. The types of activity that may be of benefit in preventing colon cancer are largely unknown. Studies analysing dose-response relationships suggest that more intense activity may confer greater protection against the risk of colon cancer than less intense activity.5 ,6
The primary postulated mechanism is that physical activity reduces intestinal transit time which would limit the time of contact between the colon mucosa and cancer promoting contents. Enhancing intestinal transit time may indirectly affect the risk of colon cancer by lowering secondary bile acid concentration or by increasing faecal short chain fatty acids.32 ,33Other mechanisms related to colon cancer risk, such as impaired immune function, specific dietary intake (for example, large amounts of alcohol or fat, low amounts of dietary fibre), or an increase in body mass index, insulin resistance, prostaglandin and triglycerides levels, body iron stores, and/or free radical scavenging enzyme activity, can be altered favourably by physical activity.32
Several studies have been published on the relationship between physical activity and cholelithiasis. While early studies yielded controversial results, most of the later studies suggest a protective effect of physical activity.8 ,9 ,34Many of the earlier studies in particular are hampered by several methodological drawbacks: no control for potential risk factors other than age (for example, body weight or diet), small sample sizes, limited methods for physical activity assessment, and low variability in physical activity among subjects. Furthermore, it cannot be ruled out that in some studies the level of physical activity was already reduced as a result of the onset of disease.
Two recent large prospective studies8 ,9 showed a relative risk (RR) of cholelithiasis of 0.63 in men and 0.69 in women when comparing the most active with the most inactive subjects, whereas sedentary lifestyle (watching television or sitting) resulted in an increased risk (RR 1.11–3.32). A clear dose-response relationship was observed, independent of several potential risk factors, strongly suggesting that (symptomatic) cholelithiasis can be prevented by physical activity, even beyond its benefit for control of body weight or diet.
The mechanisms by which physical activity may influence the pathogenesis of gall stones are poorly understood but decreased biliary cholesterol secretion, and enhanced gall bladder and colonic motility, all known to be important for gall stone formation at rest,33 are postulated. Moreover, many factors which are related to an increased risk of cholesterol gall stone disease, such as glucose tolerance, high serum levels of insulin, triglycerides, and various gall bladder regulatory hormones such as cholecystokinin, and low serum levels of high density lipoprotein cholesterol, are favourably altered by physical activity.33 ,35 ,36
The only study which examined gastrointestinal haemorrhage and physical activity was a prospective cohort study with three years of follow up in 8205 elderly subjects.37Only severe gastrointestinal haemorrhage was investigated. Physical activity was measured by self reported frequency of walking, gardening, or vigorous physical activity (resulting in sweating) three years before the study baseline. A summary variable for the three activities was also calculated. For those participants doing the activity at least three times per week, RR was significantly lower for walking (0.6) and for the summary variable (0.7) in comparison with sedentary subjects, independent of several other risk factors such as age, sex, mobility, body mass index, or health status. The RR for gardening (0.8) and vigorous physical activity (0.7) was not significantly lower. The authors hypothesised that a relatively increased gastrointestinal blood flow in physically active subjects reduced the risk of gastrointestinal haemorrhage.37 One should keep in mind that the findings of this study were restricted to severe haemorrhage in elderly subjects and that no data are available for less severe forms of haemorrhage or for younger subjects.
INFLAMMATORY BOWEL DISEASE
A limited number of studies have investigated the preventive effect of physical activity on Crohn's disease (CD) or ulcerative colitis (UC). Sonnenberg38 was the first to show a beneficial effect of physical activity by comparing incidence rates of inflammatory bowel disease among several occupations in 12 014 individuals. Sedentary and physically less demanding occupations were associated with a higher risk of inflammatory bowel disease than physically demanding occupations. As in other occupational cross sectional studies, this study may have been hampered by limitations, for instance, less demanding occupations may have attracted the chronically ill and previous occupations were not recorded. Persson and colleagues39confirmed this association in a case control study in patients with CD, but not with UC. They found RR values of 0.6 and 0.5 for weekly and daily exercise, respectively. Since then, inconsistent results have been obtained: one case control study40showed a reduced risk, both for physically active CD and UC patients, while another study in CD patients did not.41 Postulated underlying mechanisms were the stress reducing effects of physical activity as well as changes in local neuro-immuno-endocrine effects.42
While the preventive effect of physical activity remains inconclusive, it has become clear that physical activity is not harmful for patients with inflammatory bowel disease,42despite acute exercise related responses, such as increased serum malondialdehyde levels and activated neutrophils.43
Nevertheless, physical activity should be promoted as these patients have muscle weakness and are at risk of osteoporosis.44 This risk is especially high with glucocorticoid medication, which causes muscle atrophy and weakness, osteoporosis, and osteopenia.42 In addition, physical activity may reduce disease activity and improve physical health, general well being, perceived stress, and quality of life.42
A possible role of physical activity in reducing the risk of diverticular disease was suggested by observations of Manousos and colleagues45 who found that diverticular disease was more prevalent among subjects with sedentary occupations than in more active occupations. Recently, Aldoori and colleagues7 also observed an inverse relationship with physical activity in a prospective cohort of 47 678 American men during four years of follow up. After adjustment for several factors such as dietary fibre intake, overall physical activity reduced the risk of symptomatic diverticular disease (for highest versus lowest extremes, RR=0.63). Risk reduction was stronger for vigorous activity (RR=0.60) than for non-vigorous activity (RR=0.93, NS). Several specific activities reduced the risk of diverticular disease but only for the combination of jogging and running was the inverse relationship statistically significant. An increase in colonic motor activity via hormonal, vascular, and mechanical aspects, leading to a reduction in colonic transit time, was postulated as an underlying mechanism.7
Several cross sectional studies have shown an inverse independent relationship between constipation and physical activity (for example, see Everhart and colleagues,10 Donald and colleagues,46 and Kinnunen47). Also, two case control studies48 ,49showed that the defecation pattern of runners was “better” (less firm stools, higher defecation frequency, higher stool weight) than in inactive controls, a finding which could not be confirmed by the study of Klauser and colleagues.50
Inconsistent results were observed in patients who participated in a physical activity programme. Meshkinpour and colleagues48 studied eight constipated patients during four weeks of light physical activity (walking five days a week for one hour a day). The distance walked each day progressively increased from 2.9 to 5.2 km, but constipation did not change. Both Karam and Nies51 and Resende and colleagues52 showed an improvement in defecation pattern and reduced laxative use after a combined physical activity/nutrition programme in elderly patients but the effect of physical activity alone was not studied. In all of these studies physical activity was of relatively low intensity.
Thus the effect of physical activity on constipation seems likely, but has not been proved. Underlying mechanisms are unclear but a favourable effect on colonic motility, decreased blood flow to the gut, biomechanical bouncing of the gut during running, compression of the colon by abdominal musculature, and increased fibre intake as a result of increased energy expenditure have all been reported.53 ,54
As mentioned above, the mechanisms by which exercise and physical activity influence the gastrointestinal tract are poorly understood although decreased gastrointestinal blood flow, increase in gastrointestinal motility, increased mechanical bouncing, and neuro-immuno-endocrine alterations are postulated.11-13However, most of the described mechanisms have only been investigated after acute bouts of exercise. Whether or not these mechanisms are predictive of the long term effects of physical activity remains to be elucidated.
GASTROINTESTINAL BLOOD FLOW
During exercise, blood will primarily be shunted to the skin and exercising muscles at the expense of the gastrointestinal tract. Rowell and colleagues17 found a 60–70% decrease in splanchnic blood flow in humans exercising at 70% of maximal oxygen consumption (V˙O2max). At maximal exercise intensity, splanchnic blood flow may be reduced by about 80%.55
Ischaemic damage has been proposed as a causal mechanism of gastrointestinal bleeding during and after exercise. Although gastrointestinal blood loss is transient, increased faecal α1 antitrypsin and lysozyme concentrations after exercise have been found, indicative of local mucosal damage with an inflammatory response.21 Transient post-exercise lesions from the stomach to the colon have been observed in athletes and the histological picture indicates ischaemic damage.3 ,14 Critical ischaemic levels may be reached under extreme exercise conditions when hyperthermia, hypohydration, hypoglycaemia, hypoxia, or a combination of these factors are present.13 In addition, changes in blood viscosity, erythrocyte deformability, and aggregability during exercise56 may further compromise local blood flow. Epithelial cells become deprived of metabolic substrates leading to necrosis and mucosal bleeding.
In theory, critical ischaemic levels and accumulation of metabolic waste products may induce malabsorption, hypersecretion, and increased gastrointestinal permeability with endotoxaemia. Findings for the effect of exercise on absorption and acid secretion are inconsistent: both a decrease or no change in absorption57 and acid secretion58 ,59 have been reported. Exercise induced increases in intestinal permeability and mild leakage of endotoxins into the portal circulation have only been found at higher intensities.60 ,61
GASTROINTESTINAL MOTILITY AND BILE SALT METABOLISM
Data on the effect of exercise and physical activity on gastrointestinal motility are scarce and mostly indirectly obtained and limited to acute exercise.20 ,62 ,63The direct effect of exercise on gastrointestinal motility has been hypothesised to explain gastrointestinal symptoms such as heartburn (gastro-oesophageal reflux), vomiting, gastrointestinal cramps, urge to defecate, and diarrhoea.13
Gastric emptying of liquids and solids appears to be unchanged or slightly accelerated at lower exercise intensities. Above 70%V˙O2max, gastric emptying is certainly delayed.18 ,59 ,64The long term effects of gastric emptying are inconsistent: Carrio and colleagues65 found that basal gastric emptying was faster in runners than in controls while Rehreret al found no difference.64
Up to now two studies have focused on the effects of exercise and physical activity on gall bladder motility. An acute bout of aerobic exercise increased gall bladder ejection fraction slightly but insignificantly in a healthy non-obese female population.66 Also, the effect of physical activity was insignificant. In 27 obese female subjects who exercised (45 minute brisk walking five times per week) for 12 weeks, postprandial gall bladder ejection fraction increased significantly, but this increase, although larger, did not differ from that in the control group.67
Orocaecal transit time was reported to be unchanged, delayed, or accelerated20 ,53 ,68after exercise. Harris and colleagues69found that a high energy intake, which is probably correlated with intensive physical activity, was significantly correlated with fast orocaecal transit.
Studies of the effect of physical activity on colonic transit time have mostly been conducted in healthy subjects and their results are conflicting, probably due to methodological problems: four studies found accelerated transit time after physical activity53 ,54 ,70 ,71while three observed no overall effect.72-74Resende and colleagues52 observed no significant changes in gastrointestinal transit time in 12 immobile long stay patients after 12 weeks of individual sessions of massage and physical activity (five days a week, once a day).
Research on the effect of exercise and physical activity on bile salt metabolism is mainly limited to animal studies. These studies showed that physical activity of moderate intensity increased bile acid excretion.75-77Part of these changes were due to hyperphagia.75 Research on the effect of physical activity and exercise on secondary bile acid formation and absorption in humans is scarce. In obese patients, a decrease in the secondary/primary bile acid ratio has been observed after treatment with a subcaloric diet and graded physical activity.78 Sutherland and colleagues79 found that total faecal bile acid concentration was significantly lower in male distance runners than in sedentary men, while bile acid secretion was not different between groups. The lower faecal bile acid concentrations were mainly due to the higher fibre intake and subsequently higher stool weight in runners.
The frequency of most gastrointestinal symptoms is almost twice as high during running than during other endurance sports such as cycling or swimming, where up and down movements are more limited.1 ,12 The mechanical vibration of the body is more than doubled in running compared with cycling.80 The way in which this bouncing of the gut affects gastrointestinal function is still unknown.12 ,53
Many hormones associated with gastrointestinal function at rest (secretion, absorption, and motility) alter during exercise, in terms of plasma concentrations: cholecystokinin, vasoactive intestinal peptide, secretin, pancreatic polypeptide, somatostatin, peptide histidine isoleucine, peptide YY, gastrin, glucagon, motilin, catecholamines, endorphins, and prostaglandins.13 ,81 ,82However, direct proof that these hormones alter gut function during exercise or induce exercise related gastrointestinal symptoms is lacking.
With respect to immune function, the current opinion is that physical activity of moderate intensity may protect against infections by inducing changes in the activity of macrophages, natural killer cells, lymphokine activated killers cells, neutrophils, and regulating cytokines.32 Severe exercise, however, can result in a transient reduction in natural killer cells and production of free radicals, which temporarily increases the risk of infection.32 ,83Knowledge of optimal intensity and duration of physical activity for an optimal immunomodulating effect in athletes and in patients with gastrointestinal related diseases is lacking.
Strenuous exercise may induce gastrointestinal symptoms such as heartburn or diarrhoea, which may deter people from participating in physical activity. Although many symptoms are acute and transient and do not hamper the athlete's health, repeated gastrointestinal bleeding during training and competition may occasionally lead to iron deficiency and anaemia. However, these and other symptoms can often be prevented with appropriate precautions.
Physical activity, mostly performed at a relatively low intensity, may also have protective effects on the gastrointestinal tract. There is strong evidence that physical activity reduces the risk of colon cancer. Less convincing evidence is found for cholelithiasis and constipation. The putative relationship between the incidence of these gastrointestinal diseases/symptoms and amount of physical activity is shown in fig 1. Physical activity may reduce the risk of diverticulosis, gastrointestinal haemorrhage, and inflammatory bowel disease, although up to now there has been little research to substantiate this. Physical activity does not interfere with the healing process in inflammatory bowel disease and will probably not reduce the risk of rectal and gastric cancer.
Future research on exercise associated digestive processes in health or disease should explore the mechanisms involved in the potential benefits and hazards of physical activity and exercise on the gastrointestinal tract.
Abbreviations used in this paper
- relative risk
- Crohn's disease
- ulcerative colitis
- maximal oxygen consumption