An acute bout of heavy exercise induces immune system responses, which are similar to those induced by infection [3]. An increase in circulating neutrophils, monocytes and natural killer (NK) cells [27,29], a catecholamine-mediated lymphocytosis [30] and a higher plasma concentration of several hormones (e.g., epinephrine, cortisol, growth hormone and prolactin) [3] can be observed. Furthermore an enhanced release of anti-inflammatory (e.g., IL-10, IL-1ra) and pro-inflammatory cytokines (e.g., TNF-α, IL-6, IL-1β, IL-8) [27] and acute phase proteins such as C-reactive protein (CRP) is induced [3]. The expression of toll-like receptors, proteins for recognizing pathogens, is reduced [31].

Immediately post-exercise or during early recovery the changes in leukocyte counts begin to return to resting levels [27], NK cell number and activity fall below pre-exercise levels [29], the lymphocytosis turns into a cortisol-induced lymphocytopenia before returning to resting values [30], and the neutrophil:lymphocyte ratio increases, which is an accepted indicator of exercise stress [32].

T-cell function and production decreases due to high stress hormone levels and exercise-induced alterations in the pro/anti-inflammatory cytokine balance [15,16], the oxidative burst (killing capacity) of phagocytic neutrophils is reduced for several hours [3], and plasma glutamine concentration may be decreased by about 20% [1]. Serum immunoglobulin (Ig) concentration remains unaffected or slightly increases, but there is a decline in salivary IgA (s-IgA) both in concentration and secretion rate [30]. These changes of immunity cell populations and functions during early recovery may lead to a higher infection risk and the above-named “Open Window” [3].

Periods of intensified training lasting for one week or more, frequently observed over the course of a competitive season or in underperformance syndrome, may result in chronically impaired immune function and increased infection risk [25,33,34]. Chronic effects of heavy exercise not only include a higher risk to URTI but also lowered numbers of leukocytes at rest compared to sedentary people, decreased neutrophil function, serum and salivary Ig concentration, NK cell number and possibly cytotoxic activity [16,35,36,37]. Several causes for impaired immune cell function due to repeated bouts of strenuous exercise are discussed [3]:

For example, Ronsen et al. [38] showed that a recovery time of 3 h between two bouts of strenuous endurance exercise results in higher levels of stress hormones and augmented immune cell dysfunction compared with 6 h of recovery between exercise bouts. Similar findings were presented by Degerstrom & Osterud [39] with a 4 h-rest interval between two consecutive bouts.

Several trials investigated the influence of pre-exercise carbohydrate fuel state on hormonal and/or immune response to endurance exercise [51,52,53,54,55,56]. In most cases participants performed a glycogen-depleting exercise (1 h cycling)—except in one study [55]—and were then set on a high (70%–77% dietary intake from CHO/8.0 g CHO/kg bodyweight (BW) per day (/day)) or low (7%–11% dietary intake from CHO/0.5 g CHO/kg BW/day) CHO diet for two to three days [51,52,53,54,55]. Costa et al. [56] allocated their subjects into a self-selected or high (12.0 g CHO/kg BW/day) CHO diet group for a 6-day period. After completing the diet, subjects had to perform a single bout of strenuous exercise—either 1 h of cycling ergometry at 70%–75% VO_2max [51,52] or at 60% Wmax followed by a time trial [53,54] or downhill running [55]. In the study of Costa et al. [56] participants had to run 1 h/day for six days in addition to their normal training regimens to create a cycle of overload training.

Depending on tested immunological markers, it was found that exercising on a high-CHO diet compared to a low-CHO diet leads to an increased [51,53] or stable [56] blood glucose level. Plasma cortisol levels may be decreased [51,53], the post-exercise glutamine level may rise [54] or stays unaffected [52]. Different effects on immune cell counts have been observed: lower numbers of neutrophils [53,55], an attenuated post-exercise leukocytosis [55], but also unaffected leukocyte counts [52], and unaffected post-exercise lymphocytopenia [51]. A high-CHO diet during times of intensified training for six days may have a favorable effect on mucosal immunity [56].

Training on low levels of CHO availability may raise the magnitude of exercise-induced immune alterations, such as higher plasma and salivary cortisol levels [52,56], decreased glutamine levels [52], higher number of circulating immune cells [51,52] and an enhanced cytokine response [54] (Table 1).

The influence on some immune variables of a CHO containing meal with different glycemic indices (GI) and glycemic loads (GL) ingested 2–3 h before endurance exercise was also tested [59,60,61]. It was found that a pre-exercise meal with a high CHO amount (65% of energy intake) may attenuate exercise-induced cytokine response, and influences leukocyte trafficking [59]. The influence of pre-exercise meals consisting of low GI foods on exercise-induced cortisol and cytokine response compared to high GI meals remains still unclear because results are inconsistent [60,61].

Although limited evidence exists, it should be highlighted that exercising in a carbohydrate-depleted state, results in higher levels of circulating stress hormones [44], greater perturbations of immune cell subsets [43] and an impaired immune function [51]. Keeping the muscle and liver glycogen stores full is therefore a crucial factor [62].

It is well established that CHO ingestion during high-intensity exercise improves athletic performance [48] and is widely practiced by athletes. Thus, effects of acute CHO ingestion on exercise-induced changes in immune function were extensively researched during the last 15 years. This section summarizes selected results from 29 placebo-controlled and/or crossover studies addressing this topic in which three [63,64,65] referred to the same subjects and exercise mode (Table 2).

A significant higher post-exercise blood glucose level in CHO supplemented groups (SUP) relative to controls (PLA) was shown in all presented studies. Due to the maintained blood glucose level, the majority of trials revealed an attenuated cortisol level, except three studies, where post-exercise cortisol levels in SUP did not differ from those in PLA [66,67,68].

Referring to Table 2, consuming a beverage delivering at least 6% CHO (1 L/h) during a minimum 1 h lasting endurance exercise of high intensity may help to attenuate exercise-induced increases of total leukocyte count and/or leukocyte subsets such as monocytes and neutrophils. Some researchers reported a lower post-exercise lymphocytosis [65,69,70,71,72,73] and a trend to attenuate lymphocytopenia during early recovery [65,74], but these findings were not confirmed by others [67,68,75,76,77,78].

Although NK cells are part of the innate immune system [79], only few attempts were made to evaluate effects of CHO ingestion during exercise on NK cells and function—with inconclusive outcomes. No significant difference between SUP and PLA in NK cell counts was shown by several investigators [67,74,76,77,78]. Contrary to this, Nieman et al. [65] reported a significant lower number of NK cells, which was confirmed by Timmons et al. [71]. In one study, cytotoxic activity of NK cells was reduced due to CHO supplementation [69].

As shown in Table 2 the effect of CHO ingestion on exercise-induced cytokine responses was investigated in 15 of the presented studies with contradictory results. It appears that acute CHO ingestion attenuates the cytokine response to prolonged exercise on particular cytokines, such as IL-6, IL-10 and IL-1ra but not on IL-8 and TNF-α. Mucosal immunity, particular saliva flow rate and s-IgA, was measured by only three investigators with no differences between SUP and PLA [68,70,80] but a higher saliva flow rate and lower s-IgA concentration in SUP compared to fluid restriction during exercise [81].

Dosage studies were done to investigate if a higher dose (12%–12.8%) of supplemented CHO compared to a lower dose (6%–6.4%) would raise the magnitude of attenuating effects on several immunological markers [72,74]. No dose-dependent differences were found and it was concluded that ingesting at least 6% CHO beverages during exercise may sufficiently attenuate hormonal and immune responses to exercise [72,74]. Cox et al. [92] examined the effects of a 28-day pre-exercise high-CHO diet (8.5 g/kg BW/day) and acute CHO supplementation (10% CHO beverage) during exercise on cytokine responses following high-intensity cycling and concluded that chronic and acute CHO consumption do not have any synergistic effects on cytokine responses. In a following study, the same group showed that consuming a CHO-containing pre-exercise meal (2.1 g CHO/kg BW) may reduce the attenuating effects of CHO ingestion during exercise (10% CHO) on cytokine responses [93].

Consumption of small amounts of CHO (1.0–1.2 g/kg BW) and protein immediately after exercise and during recovery is generally recommended to replenish body glycogen stores [49], to stimulate muscle protein synthesis [94] and to enhance training adaptations [95]. Very few trials addressing the influence of post-exercise CHO ingestion on immune variables after strenuous exercise exist. Ingestion of 1.2 g CHO/kg BW immediately post-exercise seems to have no attenuating effect during early recovery on total numbers of leukocytes and lymphocytes but prevented neutrophil degranulation after two hours of running at 75% VO_2max [96]. Plasma concentrations of IL-6 during recovery may also be unaffected when feeding 1.0 g CHO/kg BW during early recovery following cycling at 65% VO_2max to exhaustion [97].

Glutamine is the most abundant amino acid in human muscle and plasma [32]. It is a major fuel for leukocytes and lymphocytes [104] and plays an important role in protein synthesis, cytokine production and macrophage function [98]. Prolonged exercise is associated with a decreased plasma glutamine concentration by about 20% [1] and it has been hypothesized that such a substantial fall may directly lead to immunodepression (“glutamine hypothesis”) [32] and a higher risk for URTI [105].

Due to the attractiveness of this theory and the proven beneficial effects of glutamine in some clinical situations [106], glutamine and its effects on exercise related immune parameters has received much attention [107]. To date only one study has demonstrated a prophylactic effect of glutamine supplementation on the incidence of URTI symptoms [108]. It was reported that a significant lower incidence of URTI symptoms (32%) occurred in the 7-day period following a marathon-type event in the glutamine-supplemented group of runners (5 g glutamine in 330 mL water) compared with the placebo group.

Otherwise the majority of following studies have failed to confirm initial findings or to show beneficial effects when supplementing glutamine to maintain glutamine levels during exercise on various immune parameters, such as s-IgA levels [109,110,111], post-exercise IL-6 levels [112,113], acute phase proteins [112], lymphocyte and neutrophil counts [114] and post-exercise leukocytosis and neutrophil function [115]. Therefore, investigators have not been able to verify a direct link between decreased plasma glutamine levels and immune system changes induced by prolonged exercise [116].

From a practical point of view doses in excess of 5 g glutamine have to be ingested every 30–60 min during exercise to elevate plasma glutamine concentration [105], which may not be feasible in everyday training regimens. As shown by Bacurau et al. [71] it may be possible to prevent the exercise-induced reduction of plasma glutamine concentration by delivering an adequate amount of carbohydrate during exercise.

Although the BCAAs leucine, isoleucine and valine are known to have beneficial effects on reducing exercise-induced muscle damage [117] little is known about their effects on exercise immune function. Animal feeding and in vitro studies showed that BCAAs are necessary for efficient immune function [118], because they are used directly for protein synthesis and cytokine activation [98] or glutamine synthesis [32].

Some studies investigated the effects of pre-exercise BCAA ingestion on plasma glutamine levels and other immune variables. BCAA supplementation (6 g/day) for 30 days and an additional 3 g-dose 30 min before a triathlon inhibited exercise-induced plasma glutamine fall and modified the cytokine response to exercise [119]. Interestingly a 34% decrease in reported symptoms of infection in the BCCA supplemented group compared to PLA was observed [119]. Similar outcomes were shown in a following study comparing the effects of the same supplementation regimen on immune response in triathletes and runners [120]. However some argued that the study design was complicated and different between subject groups making the interpretation of results difficult [118,121] and thereforethese findings need to be confirmed with more controlled studies [32].

Few studies have evaluated the effects of a high-fat diet (40%–62% dietary fat/day) compared to a low-fat diet (15%–19% dietary fat/day) on several aspects of post-exercise immunity [124,125,126,127]. Mainly no significant differences between the high- and low-fat diets on post-exercise lymphocyte cell counts and lymphocyte subsets [126], neutrophils and other leukocyte subsets [124] and cytokine response [124,127] were found. However, significant higher pre- and post-exercise cortisol levels [125] and decreased NK cell activity in a fat-rich diet compared to a low-fat diet [126] were shown. Some investigators argued that training on a very low-fat diet (15% dietary fat/day) may lead to an increased pro-inflammatory cytokine production [124] or an overall compromised immune function due to a negative energy balance [125] and a possible deficiency of essential micronutrients (e.g., vitamin E) [32].

The essential Omega-3 (n-3) polyunsaturated fatty acids (PUFA) eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), both found in oily fish and fish oils, are strong anti-inflammatory agents. Amongst other things they suppress the production of arachidonic acid, prostaglandins, and leucotrienes that modulate the production of pro-inflammatory cytokines [122,128,129]. Despite their beneficial health-related characteristics, limited evidence addressing exercise-related anti-inflammatory effects from n-3 PUFA supplementation exists.

Supplementation protocols varied considerably between trials and daily dosage ranged from 1.3–2.2 g EPA and 0.3–2.2 g DHA during a 4- to 6-week period before strenuous exercise [130,131,132,133,134]. Mainly no effects on post-exercise inflammatory variables or markers of oxidative stress were shown. Slight effects on cytokine milieu were revealed in only one trial when EPA and DHA were supplemented alone [130] or combined with lycopene [132]. Although dietary mixes with EPA and DHA may be beneficial in clinical trials [135], a recent study was not able to show a marked influence on post-exercise immune variables when EPA and DHA (400 mg each) were combined with other dietary immunostimulants, such as quercetin [136]. Interestingly there is a wide variance in the EPA:DHA ratio used in the presented studies, ranging from 1:4 [132] to 1:1 [133,136] and 2:1 [134] up to 4–5:1 [130,131] although general guidelines suggest an EPA:DHA ratio of 2:1 for athletes [129].