key: cord-0986821-hqlh498d
authors: NIEMAN, DAVID C.
title: Exercise and Immunity: Clinical Studies
date: 2007-09-28
journal: Psychoneuroimmunology
DOI: 10.1016/b978-012088576-3/50037-x
sha: 1274115284c8ac15e10b66b10474cb2466315929
doc_id: 986821
cord_uid: hqlh498d

nan

Exercise immunology is a relatively new area of scientifi c endeavor, with 80% of articles published during the past decade (Nieman, 2003) . Growing evidence indicates that physical activity does infl uence immune function and as a consequence risk of certain types of infection, in particular the most common of all, upper respiratory tract infections (URTI). This chapter will summarize recent investigations showing that in contrast to moderate physical activity, prolonged and intensive exertion causes numerous negative changes in immunity and an increased risk of URTI.

physically active individuals, several potential immunological effects may be contributing factors. These include exercise-induced immune modulation of cytokines; activation and changes in signal transduction of natural killer (NK) cells, macrophages, neutrophils, and tumor-infi ltrating lymphocytes; changes in the expression of cell adhesion molecules; and alterations in prostaglandins. Accumulating evidence indicates that exercise training causes changes in pro-and antiinfl ammatory cytokine levels and activity that may infl uence both the prevention and treatment of certain types of carcinomas (Allgayer et al., 2004) . This area of research endeavor is highly complex, and more research is needed.

Immune senescence or age-associated immune deficiency appears to be partly responsible for the affl ictions of old age. Elderly persons are more susceptible to many infections, autoimmune disorders, and cancers when compared with younger adults. A new and growing area of research endeavor is the study of the relationship between physical activity and immune senescence. One study of adults over the age of 65 years showed that a 10-month exercise program enhanced the antibody titer response to an infl uenza immunization , thus showing that exercise can have some immediate benefi ts for the elderly. Regular physical activity may also attenuate the age-related decrease in T lymphocyte function (Smith et al., 2004) . Very few studies have been conducted in this area, but the available data taken together suggest that exercise training may need to be long term and of suffi cient volume to induce changes in body weight and fi tness before any change in immunity can be expected in old age (Kohut and Senchina, 2004; Nieman et al., 1993) . In other words, because the aging process is so dominant in old age, long-term physical activity combined with leanness and other positive lifestyle habits may be necessary before immune function is enhanced.

Type 2 diabetes and cardiovascular disease are associated with chronic low-grade systemic infl ammation. During exercise, IL-6 is produced by muscle fi bers and stimulates the appearance in the circulation of other anti-infl ammatory cytokines such as IL-1ra and IL-10 (Petersen and Pedersen, 2005) . IL-6 also inhibits the production of the pro-infl ammatory cytokine TNF-alpha and stimulates lipolysis and fat oxidation. With weight loss from energy restriction and exercise, plasma levels of IL-6 fall, skeletal muscle TNF-alpha decreases, and insulin sensitivity improves (Ferrier et al., 2004; Ryan and Nicklas, 2004) . Thus, IL-6 release from the exercising muscle may help mediate some of the health benefi ts of exercise including metabolic control of type 2 diabetes (Petersen and Pedersen, 2005) . The exercise-induced cytokine links between adipose and muscle tissues clearly warrant further study (Tomas et al., 2004) .

Upper respiratory tract infections are the most frequently occurring illnesses in humans worldwide. More than 200 different viruses cause the common cold, and rhinoviruses and coronaviruses are the culprits 25-60% of the time. The U.S. Centers for Disease Control and Prevention has estimated that more than one billion URTIs occur annually in the United States, a leading cause of lost school and work days. The average person has two or three respiratory infections each year, with young children suffering six to seven.

Among elite athletes and their coaches, a common perception is that heavy exertion lowers resistance and is a predisposing factor to URTI. In a 1996 survey by the Gatorade Sports Science Institute of 2,700 high school and college coaches and athletic trainers, 89% checked "yes" to the question, "Do you believe overtraining can compromise the immune system and make athletes sick?" (personal communication, Gatorade Sports Science Institute, Barrington, IL).

Conversely, there is also a common belief among fi tness enthusiasts that regular exercise confers resistance against infection. In a survey of 170 non-elite marathon runners (personal best time, average of 3 hours 25 minutes) who had been training for and participating in marathons for an average of 12 years, 90% reported that they defi nitely or mostly agreed with the statement that they "rarely get sick" (unpublished observations). A survey of 750 masters athletes (ranging in age from 40 to 81 years) showed that 76% perceived themselves as less vulnerable to viral illnesses than their sedentary peers .

Although the relationship between exercise and URTI and other types of infections has been explored since early in the twentieth century (Baetjer, 1932) , the number of well-designed epidemiological and exercise training experimental trials on humans is still small, limiting our understanding of this important topic (Nieman, 1997a (Nieman, , 2003 .

Data from animal studies have been diffi cult to apply to the human condition but in general have supported the fi nding that one or two periods of exhaustive exercise following inoculation lead to a more frequent appearance of infection and a higher fatality rate (but results differ depending on the pathogen, with some more affected by exercise than others) (Friman et al., 1995; Pedersen and Bruunsgaard, 1995) . Davis et al. (1997) , for example, exposed mice to rest, 30 minutes of moderate exercise, or 2.5-3 hours of exhaustive exercise following intranasal infection with the herpes simplex virus (HSV-1). Mice exercised to fatigue had a greater overall mortality during a 21-day period than did controls or moderately exercised mice.

In humans, it is well established that various measures of physical performance capability are reduced during an infectious episode (Friman et al., 1995) . Several case histories have been published demonstrating that sudden and unexplained deterioration in athletic performance can be traced to either recent URTI or subclinical viral infections that run a protracted course. In some athletes, a viral infection may lead to a debilitating state known as "post-viral fatigue syndrome" (Maffulli et al., 1993; Parker et al., 1996) . The symptoms include lethargy, easy fatigability, and myalgia, and can persist for several months.

Together, these data imply that there is a relationship between exercise and infection, and that heavy exertion may suppress various components of immunity. Research data on the resting immunity of athletes and non-athletes, however, are limited and present a confusing picture at present (Nieman, 1997b) . For example, the few studies available suggest that the innate immune system responds differentially to the chronic stress of intensive exercise, with natural killer cell activity tending to be enhanced while neutrophil function is suppressed (Nieman et al., 1995b (Nieman et al., , 1999 Pyne et al., 1995; Smith and Pyne, 1997) . The adaptive immune system (resting state) in general seems to be largely unaffected by athletic endeavor.

Each acute bout of cardiorespiratory endurance exercise leads to transient but signifi cant changes in immunity and host defense (Gabriel and Kindermann, 1997; Hoffman-Goetz and Pedersen, 1994; Nieman et al., 1997b Nieman et al., , 2002a Nieman et al., , 2003a Nieman et al., , 2004 Pedersen and Brunsgaard, 1995) . Natural killer cell activity, various measures of T-and B-cell function, upper airway neutrophil function, and salivary IgA concentration have all been reported to be suppressed for at least several hours during recovery from prolonged, intense endurance exercise (Bruunsgaard et al., 1997; Gabriel and Kindermann, 1997; Mackinnon and Hooper, 1994; Müns, 1993; Nieman et al., 1995a Nieman et al., , 1995c Nieman et al., , 2001 Nieman et al., , 2003b Shinkai et al., 1993) . During this "open window" of decreased host protection, viruses and bacteria may gain a foothold, increasing the risk of subclinical and clinical infection ( Figure 1 ).

Although this is an attractive hypothesis, no one has yet demonstrated conclusively that athletes showing the most extreme immunosuppression are those that contract an infection (Lee et al., 1992; Mackinnon et al., 1993) . In one study, salivary IgA secretion rate decreased by nearly half in a group of 155 ultramarathon runners following a 160 km race (Nieman et al., 2006) . Nearly one in four runners reported an URTI episode during the 2-week period following the race, and the decrease in sIgA secretion rate was significantly greater in these runners (54%) compared to those not reporting URTI (31%). It is doubtful, however, that sIgA output alone can be used to predict URTI at the individual athlete level. In this study, the overall predictive value for URTI was 55%, indicating that sIgA output was more useful at the group compared to the individual level, and that other factors need to be discovered and combined with sIgA before URTI risk can be predicted for individual athletes.

Understanding the relationship between exercise and infection has potential implications for public health, and for the athlete, it may mean the difference between being able to compete or performing at a subpar level or missing the event altogether because of illness. It has been proposed that the relationship between exercise and URTI may be modeled in the form of a "J" curve (Nieman, 1997a) (Figure 2 ). This model suggests that although the risk of URTI may decrease below that of a sedentary individual when one engages in moderate exercise training, risk may rise above average during periods of excessive amounts of high-intensity exercise.

The model in Figure 2 also suggests that immunosurveillance mirrors that relationship between infection risk and exercise workload. In other words, it makes sense that if regular moderate exercise lowers infection risk, it should be accompanied by enhanced immunosurveillance. On the other hand, when an athlete engages in unusually heavy exercise workloads (e.g., overtraining or a competitive endurance race event), infection risk should be related to diminished immunosurveillance.

At present, there is more evidence, primarily epidemiological in nature, exploring the relationship between heavy exertion and infection, and these data will be reviewed fi rst followed by a section on moderate exercise training and infection (Table 1) . Much more research using larger subject pools and improved research designs is necessary before this model can be wholly accepted or rejected. The epidemiological studies summarized in Table 1 all used self-reported URTI data (primarily retrospective, with two studies using 1-year daily logs). None have attempted to verify symptomatology using viral identifi cation or verifi cation by physicians. There is some concern that the symptoms reported by endurance athletes following competitive race events may refl ect those associated with an infl ammatory response rather than URTI (Castell et al., 1997; Drenth et al., 1995; Nehlsen-Cannarella et al., 1997) . Nonetheless, the data are consistent in supporting the viewpoint that heavy exertion increases the risk of URTI, while moderate exercise training is associated with a decreased risk.

Several epidemiological reports suggest that athletes engaging in marathon-type events and/or very heavy training are at increased risk of URTI (Table 1) . Nieman et al. (1990a) researched the incidence of URTI in a group of 2,311 marathon runners who varied widely in running ability and training habits. Runners retrospectively self-reported demographic, training, and URTI episode and symptom data for the 2-month period (January, February) prior to and the 1-week period immediately following the 1987 Los Angeles Marathon race. During the week following the race, 12.9% of the marathoners reported an URTI compared to only 2.2% of control runners who did not participate (odds ratio, 5.9, using a logistic regression model that controlled for various training and demographic variables). Forty percent of the runners reported at least one URTI episode during the 2-month winter period prior to the marathon race. Controlling for various confounders, it was determined that runners training more than 96 km/wk doubled their odds for sickness compared to those training less than 32 km/wk. Linde (1987) studied URTI in a group of 44 elite orienteers and 44 non-athletes of the same age, sex, and occupational distribution during a 1-year period. Athletes and controls recorded symptoms of sickness using daily logs for 1 year. The orienteers experienced signifi cantly more URTI episodes during the year in comparison to the control group (2.5 vs. 1.7 episodes, respectively). While one-third of the controls reported no URTI during the year-long study period, this applied to only 10% of the orienteers. The average duration of symptoms in the group of orienteers was 7.9 days compared to 6.4 days in the control group (NS). The control group had the expected seasonal variation with the peak incidence in winter and relatively few cases in summer, while the orienteers tended to show a more even distribution. Heath et al. (1991) also followed a cohort of runners (N = 530) who self-reported URTI symptoms daily for 1 year. The average runner in the study was about 40 years of age, ran 32 km/wk, and experienced a rate of 1.2 URTI per year. Controlling for various confounding variables using logistic regression, the lowest odds ratio for URTI was found in those running less than 16 km/wk. The odds ratio more than doubled for those running more than 27 km/wk. This study demonstrated that total running distance for a year is a signifi cant risk factor for URTI, with risk increasing as the running distance rises. Peters and Bateman (1983) studied the incidence of URTI in 150 randomly selected runners who took part in a 56 km Cape Town race in comparison to matched controls who did not run. Symptoms of URTI occurred in 33.3% of runners compared with 15.3% of controls during the 2-week period following the race, and were most common in those who achieved the faster race times. The most prevalent symptoms after the race were reported to be sore throats and nasal symptoms. Of the total number of symptoms reported by the runners, 80% lasted for longer than 3 days, suggesting an infective origin.

Several subsequent studies from this group of researchers have confi rmed this fi nding (Peters, 1990; Peters et al., 1993 Peters et al., , 1996 . During the 2-week period following the 56 km Milo Korkie Ultramarathon in Pretoria, South Africa, 28.7% of the 108 subjects who completed the race reported non-allergy-derived URTI symptoms as compared to 12.9% of controls (Peters, 1990) . In another study, 68% of runners reported the development of symptoms of URTI within 2 weeks after the 90 km Comrades Ultramarathon (Peters et al., 1993) . Using a double-blind placebo research design, it was determined that only 33% of runners taking a 600 mg vitamin C supplement daily for 3 weeks prior to the race developed URTI symptoms. The incidence of URTI was greatest among the runners who trained the hardest coming into the race (85% vs. 45% of the low-or medium-training status runners).

In subsequent research, vitamin C, but not vitamin A and E, supplements have not been found to alter URTI rates following ultramarathon competitions (Peters et al., 1996; Peters-Futre, 1997) . As summarized in Table 1 , 15.9% of runners using vitamin C supplements (500 mg/day) reported URTI during the 2-week Nieman et al. (1989) 294 California runners 2-month recall of URTI Training 42 vs. 12 km/wk associated training for race incidence; 1 week recall after with lower URTI; no effect of race winter 5, 10, 21 km races participation on URTI Peters (1990) 108 South African marathon 2-week recall of URTI incidence URTI incidence 28.7% in runners vs. runners vs. 108 live-in and duration after 56 km 12.9% in controls after 56 km race controls race Nieman et al. (1990a) 2,311 Los Angeles marathon 2-month recall of URTI Runners training ≥97 vs. <32 km/wk at runners incidence during training for higher URTI risk; odds ratio 5.9 for marathon; 1-week recall participants vs. non-participants 1 week after winter race after 42.2 km race Heath et al. (1991) 530 runners, South Carolina 1-year daily log using self-Increase in running distance positively reported, pre-coded, related to increased URTI risk symptoms Peters et al. (1993) 84 South African marathon 2-week recall of URTI incidence URTI incidence 68% in runners vs. 45% in runners vs. 73 non-runner and duration after 90 km controls after 56 km race; 33% in controls race runners using vitamin C vs. 53% of controls Nieman (1993) 170 North Carolina 1-week recall of URTI incidence URTI reported by only 3% of (unpublished data) marathon runners after summer marathon marathoners during week after summer race race Peters et al. (1996) 178 South African runners 2-week recall of URTI incidence URTI incidence 40.4% in placebo runners vs. 162 controls and duration after 90 km vs. 15.9% using vitamin C (500 mg/d race for 3 weeks). Vitamin A and E had no additional effect Castell et al. (1996) 151 endurance athletes in 1-week recall of URTI incidence URTI reported by 81% of athletes using the United Kingdom after heavy exertion placebo vs. 49% using a glutamine-based beverage Nieman et al. (2002a) 91 marathon runners, North 15-day recall of URTI incidence URTI reported by 17% of marathoners; Carolina after marathon race event post-race sIgA: protein lower in URTI vs. non-URTI group Nieman et al. (2006) 155 ultramarathoners, 2-week recall of URTI incidence URTI reported by 24% of athletes; sIgA Western U.S. 160 km after 160 km race event decrease was 54% in URTI vs. 31% in non-URTI group period following the 1993 90 km Comarades ultramarathon in comparison to 40.4% of runners using placebo, 20-26% of runners using vitamin C with vitamin E and beta carotene, and 24.4% of controls on placebo. The authors suggested that because heavy exertion enhances the production of free oxygen radicals, vitamin C, which has antioxidant properties, may be required in increased quantities (Peters et al., 1993; Peters-Futre, 1997) . Other researchers have reported reduced URTI rates following marathon race events in runners consuming beverages containing glutamine, an amino acid used by various cells of the immune system (Castell et al., 1996 (Castell et al., , 1997 .

Predicting what runners will obtain URTI following an endurance race event from demographic and immune function parameters has proven to be a difficult task. Table 1 summarizes two studies that showed that URTI was more likely in runners experiencing low salivary IgA output during and after competing in a 160 km race event (Nieman et al., 2002a; 2006) . URTI risk following a race event may depend on the distance, with an increased incidence conspicuous only following marathon or ultramarathon events. For example, Nieman et al. (1989) were unable to establish any increase in prevalence of URTI in 273 runners during the week following 5 km, 10 km, and 21.1 km events as compared to the week before. URTI incidence was also measured during the 2 winter-month period prior to the three races, and in this group of recreational runners, 25% of those running 25 or more km/wk (average of 42 km/wk) reported at least one URTI episode, as opposed to 34% training less than 25 km/wk (average of 12 km/wk) (p = 0.09). These fi ndings suggest that, in recreational running, an average weekly distance of 42 versus 12 km is associated with either no change in or even a slight reduction of URTI incidence. Further, they suggest, that racing 5 km to 21.1 km is not related to an increased risk of sickness during the ensuing week.

URTI rates after endurance races may also vary depending on the season. As summarized in Table 1 , only 3% of marathon runners reported URTI symptoms during the week after the July 1993 Grandfather Mountain Marathon (Boone, NC).

Together, these epidemiological studies imply that heavy acute or chronic exercise is associated with an increased risk of URTI. The risk appears to be especially high during the 1-or 2-week period following marathon-type race events. Among runners varying widely in training habits, the risk for URTI is slightly elevated for the highest distance runners, but only when several confounding factors are controlled.

Thus, it makes sense that URTI risk may be increased when the endurance athlete goes through repeated cycles of unusually heavy exertion; has been exposed to novel pathogens; and has experienced other stressors to the immune system including lack of sleep, severe mental stress, malnutrition, or weight loss. A 1-year retrospective study of 852 German athletes showed that risk of URTI was highest in endurance athletes who also reported signifi cant stress and sleep deprivation (Konig et al., 2000) . In other words, URTI risk is related to many factors, and when brought together during travel to important competitive events, the athletes may be unusually susceptible.

The majority of endurance athletes, however, do not report URTI after competitive race events. For example, only 1 in 7 marathon runners reported an episode of URTI during the week following the March 1987 Los Angeles Marathon, compared to 2 in 100 who did not compete (Nieman et al., 1990a) . When athletes train hard, but avoid overreaching and overtraining, URTI risk is typically unaltered. For example, during a 2.5month period (winter/spring) in which elite female rowers trained 2-3 hours daily (rowing drills, resistance training), incidence of URTI did not vary significantly from that of non-athletic controls (Nieman et al., 2000a) .

What about the common belief that moderate physical activity is benefi cial in decreasing URTI risk? Very few studies have been carried out in this area, and more research is certainly warranted to investigate this interesting question.

At present, there are few published epidemiological reports that have retrospectively or prospectively compared incidence of URTI in large groups of moderately active and sedentary individuals. A 1-year epidemiological study of 547 adults demonstrated a 23% reduction in risk of URTI in those engaging in regular versus irregular moderate-to-vigorous physical activity (Matthews et al., 2002) . In healthy elderly subjects, URTI symptomatology during a 1-year period was inversely related to energy expended during moderate physical activity (Kostka et al., 2000) . In Project PRIME, a randomized clinical trial that investigated interventions to increase physical activity, the odds ratio for reporting URTI symptoms was 0.50 (95% C.I., 0.28 to 0.91) among participants who engaged in a minimum of 150 minutes per week of moderate and vigorous activity compared with less active participants (Strasner et al., 2001) .

Three randomized experimental trials have provided important data in support of the viewpoint that moderate physical activity may reduce URTI symptomatology (Table 2 ). In one randomized, controlled study of 36 women (mean age 35 years), exercise subjects walked briskly for 45 minutes, 5 days a week, and experienced one-half the days with URTI symptoms during the 15-week period compared to that of the sedentary control group (5.1 ± 1.2 vs. 10.8 ± 2.3 days, p = 0.039) (Nieman et al., 1990b) .

The effect of exercise training (fi ve 45-minute walking sessions/week at 60-75% maximum heart rate) and/or moderate energy restriction (4.19-5.44 MJ or 1,200-1,300 kcal per day) on URTI was studied in non-obese, physically active women (N = 30) and obese women (N = 91, body mass index 33.1 ± 0.6 kg/m 2 ) randomized to one of four groups: control, exercise, diet, exercise and diet (Nieman et al., 1998a ). All subjects self-reported symptoms of sickness in health logs using a pre-coded checklist. Energy restriction had no signifi cant effect on URTI incidence, and subjects from the two exercise groups were contrasted with subjects from the two non-exercise groups. The number of days with symptoms of URTI for subjects in the exercise groups was reduced relative to the non-exercise groups (5.6 ± 0.9 and 9.4 ± 1.1 sickness days, respectively), similar to that of the non-obese controls (4.8 ± 0.9). Figure 3 summarizes the combined data set from these two training studies.

In a study of elderly women, the incidence of the common cold during a 12-week period in the fall was measured to be lowest in highly conditioned, lean subjects who exercised moderately each day for about 1.5 hours (8%) (Nieman et al., 1993) . Elderly subjects who walked 40 minutes, 5 times/week had an incidence of 21%, as compared to 50% for the sedentary control group (X = 6.36, p = 0.042). These data suggest that elderly women not engaging in cardiorespiratory exercise are more likely than those who do exercise regularly to experience an URTI during the fall season.

During moderate exercise or vigorous exercise that incorporates rest intervals, several positive changes occur in the immune system (Nehlsen-Cannarella et al., 1991; Nieman, 2000; Nieman and Nehlsen-Cannarella, 1994; Nieman et al., 2005a) . Stress hormones, which can suppress immunity, and pro-and anti-infl ammatory cytokines, indicative of intense metabolic activity, are not elevated during moderate exercise. Although the immune system returns to preexercise levels very quickly after the exercise session is over, each session represents a boost in immune surveillance that appears to reduce the risk of infection over the long term.

Although public health recommendations must be considered tentative, the data on the relationship between moderate exercise, enhanced immunity, and lowered risk of sickness are consistent with guidelines Nieman et al. (1993) 42 elderly women (30 Daily logs using self-reported, Incidence of URTI 8% in athletes; inactive, 12 athletes), pre-coded, URTI symptoms for inactives randomized to 12 weeks North Carolina 12 weeks during fall season walking vs. controls, 21% in walkers, 50% in sedentary controls Nieman et al. (1998a) 90 overweight and 30 Daily logs using self-reported, Number of days with URTI symptoms, normal-weight women pre-coded URTI symptoms for 9.4 ± 1.1, 5.6 ± 0.9, and 4.8 ± 0.9 in 12 weeks during winter season overweight walkers, overweight controls, and normal-weight controls

The number of URTI symptom days was decreased by approximately half through a walking program (5 days per week, 45 minutes per session, for 15 weeks).

urging the general public to engage in near-daily brisk walking.

Endurance athletes are often uncertain of whether they should exercise or rest during an infectious episode. There are few data available in humans to provide defi nitive answers. Most clinical authorities in this area recommend that if the athlete has symptoms of a common cold with no constitutional involvement, then regular training may be safely resumed a few days after the resolution of symptoms.

Mild exercise during sickness with a common cold does not appear to be contraindicated. Weidner et al. (1997) , for example, have shown that rhinovirus-caused upper respiratory illness does not impair short duration submaximal or maximal exercise performance. In addition, moderate exercise training did not infl uence URTI symptomatology (Weidner et al., 1998; Weidner and Schurr, 2003) . However, it should be cautioned that rhinoviruses account for only 40% of URTI, and further research is needed with other pathogens to determine their relationship to exercise. Some clinicians feel that if there are symptoms or signs of systemic involvement (fever, extreme tiredness, muscle aches, swollen lymph glands, etc.), then 2-4 weeks should probably be allowed before resumption of intensive training (Friman et al., 1995; Nieman, 1997a) .

For elite athletes who may be undergoing heavy exercise stress in preparation for competition, several precautions may help them reduce their risk of URTI. Considerable evidence indicates that two other environmental factors-improper nutrition and psychological stress-can compound the negative infl uence that heavy exertion has on the immune system (Nieman, 1997a; Shephard and Shek, 1995) . Based on current understanding, the athlete is urged to eat a well-balanced diet, keep other life stresses to a minimum, avoid overtraining and chronic fatigue, obtain adequate sleep, and space vigorous workouts and race events as far apart as possible.

Immune system function may be suppressed during periods of rapid weight loss, so when necessary, the athlete is advised to lose weight slowly during noncompetitive training phases (Nieman et al., 1998a; Shade et al., 2004) . Cold viruses are spread by both personal contact and breathing the air near sick people. Therefore, if at all possible, athletes should avoid being around sick people before and after important events. If the athlete is competing during the winter months, a fl u shot is recommended.

Nutrition impacts the development of the immune system, both in the growing fetus and in the early months of life. Nutrients are also necessary for the immune response to pathogens so that cells can divide and produce antibodies and cytokines. Many enzymes in immune cells require the presence of micronutrients, and critical roles have been defi ned for zinc; iron; copper; selenium; and vitamins A, B6, C, and E in the maintenance of optimum immune function.

Although endurance athletes may be at increased infection risk during heavy training or competitive cycles, they must exercise intensively to contend successfully. Athletes appear less interested in reducing training workloads, and more receptive to ingesting nutrient supplements that have the potential to counter exercise-induced infl ammation and immune alterations (Nieman, 2001; Shephard and Shek, 1995) .

The infl uence of a growing list of nutritional supplements on the immune and infection response to intense and prolonged exercise has been assessed. Supplements studied thus far include zinc, dietary fat, plant sterols, antioxidants (e.g., vitamins C and E, betacarotene, N-acetylcysteine, and butylated hydroxyanisole), glutamine, and carbohydrate. Except for carbohydrate beverages, none of these supplements has emerged as an effective countermeasure to exercise-induced immune suppression. Antioxidants and glutamine have received much attention, but the data thus far do not support their role in negating immune changes after heavy exertion. At this point, athletes should eat a varied and balanced diet in accordance with the Food Guide Pyramid and energy needs, and be assured that vitamin and mineral intake is adequate for both health and immune function.

Research during the 1980s and early 1990s established that a reduction in blood glucose levels was linked to hypothalamic-pituitary-adrenal activation, an increased release of adrenocorticotrophic hormone and cortisol, increased plasma growth hormone, and increased plasma epinephrine levels (Murray et al., 1991) . Stress hormones have an intimate link with some aspects of immune function (Figure 4) .

Several studies with runners and cyclists have shown that carbohydrate beverage ingestion plays a role in attenuating changes in immunity when the athlete experiences physiologic stress and depletion of carbohydrate stores in response to high intensity (∼75-80% VO 2max ) exercise bouts lasting longer than 90 minutes (Nehlsen-Cannarella et al., 1997; Nieman et al., 1997a Nieman et al., , 1997c Nieman et al., , 1998b Nieman et al., , 2003a . In particular, carbo-hydrate ingestion (about one liter per hour of a typical sports drink) compared to a placebo has been linked to reduced change in blood immune cell counts, and lower pro-and anti-infl ammatory cytokines. However, carbohydrate ingestion during intense and prolonged exercise is largely ineffective in countering postexercise decrements in NK and T-cell function, and salivary IgA output.

Overall, these data indicate that physiological stress to some aspects of the immune system is reduced when endurance athletes use carbohydrate during intense exertion lasting 90 minutes or more. Does this mean that endurance athletes using carbohydrate beverages during marathon-type race events will lower their risk of sickness afterwards? One study suggests this is so, but more research is needed (Nieman et al., 2002a) .

Heavy exertion increases the generation of free radicals and reactive oxygen species (ROS) through several pathways including oxidative phosphorylation, increase in catecholamines, prostanoid metabolism, xanthine oxidase, and NAD(P)H oxidase (Urso and Clarkson, 2003) . Neutrophils and macrophages migrate to the site of contraction-induced muscle damage, infi ltrate the muscle tissue, activate the release of cytokines, and produce additional ROS. Most ROS are neutralized by a sophisticated antioxidant defense system consisting of a variety of enzymes and nonenzymatic antioxidants including vitamin A, E, and C; glutathione; ubiquinone; and fl avonoids. Intensive and sustained exercise, however, can create an imbalance between ROS and antioxidants, leading to oxida-tive stress that not only causes lipid peroxidation and protein oxidation, but may also impact immune function.

Can antioxidant supplements attenuate exerciseinduced changes in immune function and infection risk? Several double-blind placebo studies of South African ultramarathon runners demonstrated that vitamin C (but not E or beta-carotene) supplementation (about 600 mg/day for 3 weeks) was related to fewer reports of URTI symptoms (Peters, 1990; Peters et al., 1993 Peters et al., , 1996 Peters and Bateman, 1983; Peters-Futre, 1997 ). This has not been replicated, however, by other research teams. Himmelstein et al. (1998) , for example, reported no alteration in URTI incidence among 44 marathon runners and 48 sedentary subjects randomly assigned to a 2-month regimen of 1,000 mg/ day of vitamin C or placebo. Most randomized, placebo-controlled studies have been unable to demonstrate that vitamin C supplements modulate immune responses following heavy exertion Nieman et al., 2000b Nieman et al., , 2002b .

Vitamin E functions primarily as a non-specifi c, chain-breaking antioxidant that prevents the propagation of lipid peroxidation. The vitamin is a peroxyl radical scavenger and protects polyunsaturated fatty acids within membrane phospholipids and in plasma lipoproteins. The effect of vitamin E supplementation on the infl ammatory and immune response to intensive and prolonged exercise is largely unstudied and equivocal. Cannon et al. (1991) found that vitamin E supplementation of 800 IU/day for 48 days attenuated endotoxin-induced IL-6 secretion from mononuclear cells for 12 days after running downhill on an inclined treadmill. Singh et al. (1999) showed no effect of vitamin E supplementation (4 days, 800 IU/day) on the increase in plasma IL-6 following a 98-minute treadmill run at 65-70% VO 2max to exhaustion. Petersen et al. (2002) reported no infl uence of vitamin E and C supplementation (500 mg and 400 mg, respectively, for 14 days before and 7 days after) on the plasma cytokine response to a 5% downhill 90-minute treadmill run at 75% VO 2max .

Two months of vitamin E supplementation at a dose of 800 IU/day α-tocopherol did not effect increases in plasma cytokines, perturbations in other measures of immunity, or oxidative stress in triathletes competing in the Triathlon World Championship race event (Nieman et al., 2004) . To the contrary, athletes in the vitamin E compared to placebo group experienced greater lipid peroxidation and increases in plasma levels of several cytokines following the triathlon. Despite these indications that vitamin E exerted prooxidant and pro-infl ammatory effects, race performance did not differ between athletes in the vitamin E FIGURE 4 Carbohydrate supplementation during exercise attenuates plasma stress hormone, blood neutrophil count, and plasma cytokine responses. and placebo groups. In general, vitamin E supplementation to counter immune suppression and oxidative stress in endurance athletes cannot be recommended.

Glutamine, a non-essential amino acid, has attracted much attention by investigators (Mackinnon and Hooper, 1996; Rhode et al., 1995 Rhode et al., , 1998 . Glutamine is the most abundant amino acid in the body, and is synthesized by skeletal muscle and other tissues. Glutamine is an important fuel for lymphocytes and monocytes, and decreased amounts in vitro have a direct effect in lowering proliferation rates of lymphocytes.

Reduced plasma glutamine levels have been observed in response to various stressors, including prolonged exercise. Since skeletal muscle is the major tissue involved in glutamine production and is known to release glutamine into the blood compartment at a high rate, it has been hypothesized that muscle activity may directly infl uence the immune system by altering the availability of this immune cell fuel substrate.

Whether exercise-induced reductions in plasma glutamine levels are linked to impaired immunity and host protection against viruses in athletes is still unsettled, but the majority of studies have not favored such a relationship (Nieman and Pedersen, 2000) . For example, in a crossover, placebo-controlled study of eight males, glutamine supplementation abolished the post-exercise decrease in plasma glutamine concentration but still had no infl uence relative to placebo on exercise-induced decreases in T-and natural killer cell function (Rhode et al., 1998) .

One problem with the glutamine hypothesis is that plasma concentrations following exercise do not decrease below threshold levels that are detrimental to lymphocyte function as demarcated by in vitro experiments. In other words, even marathon-type exertion does not deplete the large body stores of glutamine enough to diminish lymphocyte function.

In summary, there is growing evidence that prolonged intensive exercise is associated with an increased risk of URTI, in contrast to a decreased risk with moderate exercise training. Attempts have been made to alter the elevated URTI risk and suppressed immunity that have been associated with heavy exertion through chemical and nutritional means, with the most promising results thus far reported for carbohydrate supplementation. Further research is needed to establish the relationship of other nutrient supplements to changes in both immune function and host protection against URTI pathogens.

By far, the most important fi nding that has emerged from exercise immunology studies during the past 2 decades is that positive immune changes take place during each bout of moderate physical activity. Over time, this translates to fewer days of sickness with the common cold and other upper respiratory tract infections. This is consistent with public health guidelines urging individuals to engage in near-daily physical activity of 30 minutes or greater.

The future of exercise immunology is determining whether or not exercise-induced perturbations in immunity help explain improvements in other clinical outcomes such as cancer, heart disease, type 2 diabetes, arthritis, and aging. This is an exciting new area of scientifi c endeavor, and preliminary data suggest that immune changes during exercise training are one of multiple mechanistic factors.

Risk of upper respiratory tract infections can increase when athletes push beyond normal limits. The infection risk is amplifi ed when other factors related to immune function are present, including exposure to novel pathogens during travel, lack of sleep, severe mental stress, malnutrition, or weight loss.

Should the athlete exercise when sick? In general, if the symptoms are from the neck up (e.g., the common cold), moderate exercise is probably acceptable and some researchers would argue even benefi cial, while bedrest and a gradual progression to normal training are recommended when the illness is systemic (e.g., the fl u). If in doubt as to the type of infectious illness, individuals should consult a physician.

Many components of the immune system exhibit adverse change after prolonged, heavy exertion lasting longer than 90 minutes. These immune changes occur in several compartments of the immune system and body (e.g., the skin, upper respiratory tract mucosal tissue, lung, blood, and muscle). During this "open window" of impaired immunity (which may last between 3 and 72 hours, depending on the immune measure), viruses and bacteria may gain a foothold, increasing the risk of subclinical and clinical infection. Of the various nutritional countermeasures that have been evaluated thus far, ingestion of carbohydrate beverages during intense and prolonged exercise has emerged as the most effective. However, carbohydrate supplementation during exercise decreases exerciseinduced increases in plasma cytokines and stress hormones, but is largely ineffective against other immune components including natural killer and T-cell function.

Decreased interleukin-1 receptor antagonist response following moderate exercise in patients with colorectal carcinoma after primary treatment

The effect of muscular fatigue upon resistance

In vivo cellmediated immunity and vaccination response following prolonged, intense exercise

Acute phase response to exercise. II. Associations between vitamin E, cytokines, and muscle proteolysis

Some aspects of the acute phase response after a marathon race, and the effects of glutamine supplementation

Does glutamine have a role in reducing infections in athletes?

Exercise, alveolar macrophage function, and susceptibility to respiratory infection

Endurance run increases circulating IL-6 and IL-1ra but down regulates ex vivo TNF-α and IL-1β production

Randomized controlled trial of exercise and blood immune function in postmenopausal breast cancer survivors

Diet but not aerobic exercise training reduces skeletal muscle TNF-alpha in overweight humans

Infectious and lymphocytic myocarditis: epidemiology and factors relevant to sports medicine. Scand

The acute immune response to exercise: what does it mean?

Exercise and the incidence of upper respiratory tract infections

Vitamin C supplementation and upper respiratory tract infections in marathon runners

Exercise and cancer: do the biology and epidemiology correspond?

Exercise and the immune system: a model of the stress response?

Reversing age-associated immunosenescence via exercise

Upper respiratory tract infection in athletes: infl uence of lifestyle, type of sport, training effort, and immunostimulant intake

The symptomatology of upper respiratory tract infections and exercise in elderly people

Immune responsiveness and risk of illness in U. S. Air Force Academy cadets during basic cadet training

Running and upper respiratory tract infections. Scand

Temporal relationship between decreased salivary IgA and upper respiratory tract infection in elite athletes

Mucosal (secretory) immune system responses to exercise of varying intensity and during overtraining

Plasma glutamine and upper respiratory tract infection during intensifi ed training in swimmers

Post-viral fatigue syndrome. A longitudinal assessment in varsity athletes

Moderate to vigorous physical activity and risk of upper-respiratory tract infection

Effect of long-distance running on polymorphonuclear neutrophil phagocytic function of the upper airways

Responses to varying rates of carbohydrate ingestion during exercise

Carbohydrate and the cytokine response to 2.5 hours of running

The effects of acute moderate exercise on lymphocyte function and serum immunoglobulins

Exercise immunology: practical applications

Immune response to heavy exertion

Exercise effects on systemic immunity

Exercise immunology: nutritional countermeasures

Current perspective on exercise immunology

Indomethacin does not alter the natural killer cell response to 2.5 hours of running

Immune function in marathon runners versus sedentary controls

Carbohydrate ingestion infl uences skeletal muscle cytokine mRNA and plasma cytokine levels after a 3-h run

Immune and oxidative changes during and following the Western States Endurance Run

Carbohydrate supplementation affects blood granulocyte and monocyte traffi cking but not function following 2.5 hours of running

The immune response to a 30-minute walk

Vitamin C supplementation does not alter the immune response to 2.5 hours of running

Relationship between salivary IgA secretion and upper respiratory tract infection following a 160-km race

Change in salivary IgA following a competitive marathon race

Carbohydrate affects natural killer cell redistribution but not activity after running

Physical activity and immune function in elderly women

Vitamin E and immunity after the Kona Triathlon World Championship

Infl uence of vitamin C supplementation on oxidative and immune changes following an ultramarathon

Cytokine changes after a marathon race

Infectious episodes in runners before and after a roadrace

Infectious episodes in runners before and after the Los Angeles Marathon

The immune response to exercise

Immune response to two hours of rowing in female elite rowers

Immune function in female elite rowers and nonathletes

Immune response to exercise training and/or energy restriction in obese women

Infl uence of carbohydrate ingestion and mode on the granulocyte and monocyte response to heavy exertion in triathletes

The effects of moderate exercise training on natural killer cells and acute upper respiratory tract infections

Nutrition and exercise immunology

Infl uence of vitamin C supplementation on cytokine changes following an ultramarathon

Lymphocyte proliferation response to 2.5 hours of running

Chronic fatigue syndrome and the athlete

How physical exercise infl uences the establishment of infections

Altitude fails to increase susceptibility of ultramarathon runners to post-race upper respiratory tract infections

Respiratory tract infections: an epidemiological survey

Vitamin C supplementation reduces the incidence of postrace symptoms of upper-respiratory-tract infection in ultramarathon runners

Vitamin C as effective as combinations of anti-oxidant nutrients in reducing symptoms of upper respiratory tract infection in ultramarathon runners

Vitamin C, neutrophil function, and upper respiratory tract infection risk in distance runners: the missing link

Effect of vitamin supplementation on cytokine response and on muscle damage after strenuous exercise

The anti-infl ammatory effect of exercise

Effects of an intensive 12-wk training program by elite swimmers on neutrophil oxidative activity

Effect of glutamine supplementation on changes in the immune system induced by repeated exercise

Effects of glutamine on the immune system: infl uence of muscular exercise and HIV infection

Reductions in plasma cytokine levels with weight loss improve insulin sensitivity in overweight and obese postmenopausal women

Frequent intentional weight loss is associated with lower natural killer cell cytotoxicity in postmenopausal women: possible long-term immune effects

Personal health benefi ts of Masters athletics competition

Heavy exercise, nutrition and immune function: Is there a connection?

Triathlon competition induced a transient immunosuppressive change in the peripheral blood of athletes

Neuroendocrine responses to running in women after zinc and vitamin E supplementation

Exercise, training, and neutrophil function

Infl uence of age and physical activity on the primary in vivo antibody and T cell-mediated responses in men

Impact of physical activity on URTI symptoms in Project PRIME participants

Metabolic and hormonal interactions between muscle and adipose tissue

Oxidative stress, exercise, and antioxidant supplementation

Effect of a rhinovirus-caused upper respiratory illness on pulmonary function test and exercise responses

The effect of exercise training on the severity and duration of a viral upper respiratory illness

Effect of exercise on upper respiratory tract infection in sedentary subjects