Part I: Relationship among Training Load Management, Salivary Immunoglobulin A, and Upper Respiratory Tract Infection in Team Sport: A Systematic Review

Immunoglobulin A (IgA), which is the first line of defense against upper respiratory tract viruses, has been related with training load management. This article aimed to systematically identify and summarize (1) the studies that have found a relationship between training load and salivary IgA in team sports, and (2) the studies that have highlighted a relationship between IgA and upper respiratory tract infection (URTI) in team sports. A systematic review of relevant articles was carried out using two electronic databases (PubMed and WoK) until 3 October 2020. From a total of 174 studies initially found, 24 were included in the qualitative synthesis. This systematic review confirmed that lower values of IgA occurred after greater training load (intensity/volume) and congested periods. In this scenario, a low level of IgA was correlated with higher URTI, which makes training load management mandatory to healthcare avoiding immunosuppression. Therefore, physical fitness and conditioning coaches should carefully manage training load progression, avoiding high-intensity sessions in two consecutive days. In addition, they should not program high-intensity training sessions during at least the two days following competition.


Introduction
Upper respiratory tract infections (URTIs) have been extensively studied in the context of exercise [1], since they are the most prevalent illness reported in athletes [2]. In a theoretical model called the "J" curve, it is hypothesized that moderate exercise is associated with a lower risk of infection, while vigorous exercise performed with frequency or density may lead to an increase in the risk of infection by contributing to a diminishing of immunosurveillance [3]. Some stressors such as high training load, congested match schedules, travels, environmental extremes, or lifestyle are some of the main reasons to explain the exposure to a diminished immunosurveillance [4].
Among others, immunoglobulin A (IgA) is one of the main active antibodies in defending against infectious agents. IgA acts as the first line of defense, inhibiting the bacterial Healthcare 2021, 9, 366 2 of 22 and viral adhesion to epithelial cells and neutralizing bacterial toxins and viruses [5]. In particular, secretory IgA plays a determinant role in mucosal surfaces (e.g., respiratory tract) which is determinant for fighting the entry of many pathogens [6,7]. Despite a great intra-and inter-individual variability of IgA levels in athletes [8], salivary IgA seems to be an effective outcome for controlling the risk of developing URTIs in athletes [4].
Training load and training periodization can be considered some of the stressors responsible for affecting salivary IgA. However, findings in the original studies are inconsistent, and, in some cases, correlations between accumulated training load and IgA are unclear [9,10]. The evidence is also unclear in acute response, since some studies presented no significant changes in salivary IgA concentrations after a match [11][12][13], while others presented a significant reduction in salivary IgA concentrations hours after competition or heavy training [14][15][16].
In the pandemic context, in which SARS-CoV-2 presents high rates of infection by severely affecting the URT, it is determinant to identify how the first line of defense of the respiratory track (i.e., IgA) may vary on the basis of training load imposed in athletes. In fact, team sports players are still training and competing with very congested schedules, leading to accumulated loads that should be considered from an immunological point of view [17]. For that reason, aiming to identify how training load may affect the salivary IgA levels in team sports players, it is important to conduct a systematic review. So far, there has been a narrative review that analyzed the impact of sport related stressors on immunity and illness risk in team sports players [4]. However, as far we may know, no systematic review was conducted about the influence of training load on the IgA levels of team sports players. This may help to understand the consequences of training load on IgA and, ultimately, to provide information or guidelines to protect the players from URTIs.
Accordingly, the aim of this article was to systematically identify and summarize (1) the studies that have found a relationship between training load and salivary IgA in team sports, and (2) the studies that have highlighted a relationship between IgA and upper respiratory tract infection (URTI). We hypothesized that lower values of IgA are present after greater training load periods and congested schedules, making rest strategies necessary after matches and high-intensity training sessions. In addition, we hypothesized that IgA is correlated with URTI, which makes non-congested schedules mandatory.

Materials and Methods
This systematic review is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta Analyses (PRISMA) guidelines [18].

Design
A systematic search was performed by two authors to identify articles published before 3 October 2020 in PubMed and Web of Science. The following search strategy was used (sport, intervention, and population): ("team sport*" OR handball OR hockey OR basketball OR rugby OR soccer OR football OR futsal OR "indoor football") AND (salivary) AND (immunity OR "Immunoglobulin A" OR sIgA OR "sIgA secretion rate" OR "srIgA" OR "mucosal immunity" OR "upper respiratory symptom*" OR URS). Due to the high number of articles found, the present systematic review summarized all articles performed in team sports (rugby, Australian football, basketball, handball, ice hockey, futsal), while the articles carried out with soccer players were summarized in another systematic review (Part II).

Screening Strategy and Study Selection
After completion of the search, results were compared between researchers (M.R.G. and J.P.O.) to ensure that the same number of articles was found. Then, one of the authors (M.R.G.) downloaded the main data from the articles (title, authors, date, and database) to an Excel spread sheet (Microsoft Excel, Microsoft, Redmond, DC, USA) and removed duplicate records. Subsequently, the same authors screened the remaining records to verify the inclusion/exclusion criteria using a hierarchical approach. The papers were excluded when they met the exclusion criteria in Table 1. Articles that assessed the influence of another factor (e.g., nutritional intervention, recovery strategies) in IgA response or the effects of IgA in other contexts (e.g., oral health) 4 Original articles Nonoriginal research papers (i.e., systematic reviews, conferences, or meetings) 5 Articles developed with team sport athletes (rugby, basketball, futsal, Australian football, ice hockey, or handball) more than 18 years old.
Articles developed with soccer players or with non-team sport athletes

Methodological Assessment
The methodological assessment process was performed by two authors (M.R.G. and N.B.H.) using an adapted version of the STROBE assessment criteria for cross-sectional studies, finding studies eligible for inclusion. Each article was assessed on the basis of 10 specific criteria (see Table 2). Any disagreement was discussed and solved by consensus decision. Each item was evaluated using numerical characterization (1 = completed; 2 = non-completed). As suggested by O Reilly et al. [19], each study rating was qualitatively interpreted as follows: if score ≥ 7, the study was qualified as having a low risk of bias; otherwise, the study was qualified as having a high risk of bias.

Identification and Selection of Studies
A total of 174 (PubMed = 67; Web of Sciences = 107) documents were initially retrieved from the databases, of which 63 were duplicated. Thus, a total of 111 articles were downloaded. After screening the titles and abstract against criterion 1 (where applicable), as well as the full text of the remaining papers against the same criterion, two studies were excluded. From the 85 articles, 7, 14, 7, and 58 were ruled out against criteria 2, 3, 4, and 5, respectively. Therefore, 23 studies were included for the qualitative analysis ( Figure 1). multiplicity of analyses, results from similar studies, and other relevant evidence (item 9). Give the source of fundin the role of the funders for the present study and, if applicable, for the original study on which the present article is (item 10). Score: 1 = Yes; 2 = No.

Identification and Selection of Studies
A total of 174 (PubMed = 67; Web of Sciences = 107) documents were i trieved from the databases, of which 63 were duplicated. Thus, a total of 111 ar downloaded. After screening the titles and abstract against criterion 1 (where a as well as the full text of the remaining papers against the same criterion, two stu excluded. From the 85 articles, 7, 14, 7, and 58 were ruled out against criteria 2 5, respectively. Therefore, 23 studies were included for the qualitative analysis

Methodological Quality
The quality assessment of the cross-sectional studies can be found in Table 2.
A detailed description of the 23 studies regarding the sample, the stressor, IgA, and their conclusions is presented in Tables 3-5.

No.
Small sample size, and only one game was analyzed.
Low-intensity matches led to lack of change in sIgA. At rest, before the start of the match and 5-10 min after the match. sIgA before = 715.6 ± 214.8; sIgA after = 463.9 ± 154.6 µg/mL (p < 0.05).

No
Small sample size, and only 1 match was analyzed.
Simulated futsal matches induce a high level of stress, moderated by the high-magnitude internal load and by decreasing the level of SIgA.

No.
Small sample size.
No changes or differences in sIgA were noted between pre-and post-moments across all matches.
The inability to monitor the salivary parameters each week. Correlations over the 4 time points were not applied, and the sample size was small.
Initial levels of sIgA secretion were negatively and significantly correlated with the signs and symptoms of stress at week 4, which could indicate that the athletes with low levels of sIgA secretion were more susceptible to be affected by different stressors. No.
There were some congested periods which were not analyzed, specifically. Moreover, the sample size was small.
Lower concentrations of sIgA during URTI in athletes were observed, which may confirm the suppression of mucosal immunity and initiation responses to pathogenic infections by innate immunity. Table 3 is organized according to the main stressor of matches. Specifically, when the stressor was matches, they were split according to official matches [11,21,30,32,35], training matches [20,24,30,32], both official and training matches [36]. Table 4 is organized according to the main stressor of training which included training (only general training periods) [10,27,[37][38][39], training with strength/resistance [22,23,25,33].
Finally, Table 5 is organized with the last stressor considered which included periods with training and official matches [26,28,29,31].

Discussion
The aim of this systematic review was to investigate the studies that have found (i) a relationship between training load and salivary IgA (sIgA) in team sports, and (ii) a relationship between sIgA and URTI. Among the studies found, the main stressors used in studies were official matches [11,13,21,34,35], training matches [20,24,30,32], simultaneous official and training matches [36], periods of general training [10,27,[37][38][39][40], periods of training that included strength training [21,22,24,32], and simultaneous training periods and matches [25,27,28,30]. Although not all studies reported narrow relationship between IgA and training load and between IgA and URTI, a trend was found between training/competition volume/intensity and IgA.

IgA outcomes
This review found inconsistent results regarding the IgA outcomes. First, it is important to clarify that IgA was analyzed as salivary concentration or secretion. In general, sIgA concentration was measured by enzyme-linked immunosorbent assay (ELISA). The sIgA secretion rate (µg·min −1 ) was calculated by multiplying the absolute sIgA concentration by salivary flow rate (mL·min −1 ), and salivary flow rate was determined by dividing the volume of saliva collected by the duration of the sampling period.

Official Matches
Regarding the studies that analyzed IgA through official matches [11,13,20,21,34,35,41], Lindsay et al. [11] found that three professional rugby matches decreased sIgA concentrations, although not all of them were significant. The same study found that there were decreases over time when analyzing subject by subject, which led to the speculation that some players could possibly be better prepared for matches and, thus, less exposed to immune system fragility. On the one hand, some athletes showed significant levels of muscle damage, stress, and immune system suppression; on the other hand, some athletes did not present such levels. Some additional explanations could be attributed to confidence aspects, experience, or even the lack of match time. Moreover, Coad et al. [35] analyzed 16 Australian Rules Football League matches and found significant decreases in sIgA concentration after 36 h in eight of the 16 matches analyzed (range values = 110.31 ± 72.01-149.35 ± 68.50 µg/mL). In the same way, Mariscal et al. [21] found that one handball official match significantly decreased sIgA concentration (mean difference = 495.5 µg/mL (95% CI = 223.48-767.93)). In contrast, Koch et al. [34] and Moreira, Bacarau et al. [13] found that one rugby and basketball match did not present significant decreases, respectively. Some possible differences in the results could be associated with the specificities of the sport, such as technical/tactical actions, duration of the matches, intensity of the matches, and some contextual factors, such as the quality of the opponents, match location (home versus away), and match result (weather the team is winning, loosing, or drawing) that could influence the results. In addition, sIgA following acute exercise did not change as previously reported [42], but intensity of the exercise can also be a major factor affecting sIgA outcomes. Furthermore, the volume and composition of fluid consumed during the match or training and sweat rates could influence the results. Lastly, the timing of saliva sample collection may also have been a factor in the outcomes. If the collection is immediately before the training or match, it may not reflect a true resting value because players could be in a higher state of excitement, which can arouse psychological activity [42]. The same could be applied after the match. For instance, only one study did not collect immediately after the match, but 36 h after, and still found significant decreases in sIgA concentrations (pre-to post-match, 110.31 ± 72.01-136.74 ± 63.16 µg/mL) [34] while the other studies collected saliva samples 5-15 min after the end of the match [11,13,20,33].

Training Matches
When the stressor was training or friendly matches, Lindsay et al. [20] (sIgA concentration pre-to post-match 409 ± 223 to 414 ± 255 µg/mL) and Caetano Júnior et al. [24] (preto post-match values not described) found no significant differences after one rugby and handball match, respectively, while Arruda et al. [32] found significant decreases after one futsal match (pre-to post-match sIgA concentration, 715.6 ± 214.8-463.9 ± 154.6 µg/mL) and Coad et al. [30] found significant decreases after 12 and 36 h (sIgA concentrations pre-match = 290.21 ± 165.28; post 12 h = 147.20 ± 83.94; post 36 h = 113.32 ± 95.26 µg/mL) of three different Australian football matches. On one hand, the physical demands of the rugby and handball matches also included periods of recovery between efforts and relatively short durations of action, which, associated with the characteristics of friendly matches, could have influenced the outcomes in mucosal parameters; however, this was not investigated properly in both studies. As mentioned before, other justifications could be the psychological factor, once the matches were in training or simulated conditions, and the exercise intensity/duration [43]. On the other hand, a futsal training match, due to higher intensity, may be sufficient to decrease IgA. Future studies are required to better understand these findings.

Official and Training Matches
In addition, one study compared official and training matches [36]. The authors found that no significant changes in sIgA concentrations were observed across either the simulated (pre-to post-match, 494 ± 99-635 ± 137 µg/mL) or the official matches (preto post-match, 457 ± 68 552 ± 59 µg/mL). Beyond the explanations given before in an attempt to justify the findings, it is suggested that the glucocorticoid family of steroids, such as cortisol, could be responsible for controlling IgA changes [44]; however, both studies failed to prove that cortisol does play such a role in controlling IgA acute responses of exercise. Instead, they showed an inverse relationship between salivary IgA and cortisol concentrations in young adults without exercise [45].

General Training and Training with Strength/Resistance
Regarding the studies that only analyzed periods of general training [10,27,[37][38][39][40], Moreira et al. [37] found significant decreases in sIgA concentrations after a 17 day period, which was performed with maximal intensity (pre-to post-training, 541 ± 226-381 ± 111 µg/mL), but no training program was revealed. Orysiak et al. [38] also analyzed a 17 day period. During that period, three saliva collections were taken (beginning, day 9, and day 13). Higher load was imposed between the first two assessments when compared with the period between the last two assessments. As expected, it was found that the first period significantly decreased IgA (116.17 ± 51.66-113.99 ± 76.57 µg/mL), especially when compared with the last (122.84 ± 82.87 µg/mL). In contrast, Nunes, Crewther, Viveiros et al. [33] found that a 50 day period of resistance training with progressive load through the weeks decreased IgA concentrations (values not described), and Milanez et al. [23] found that 5 weeks with strength training, physical training, and technical/tactical training also decreased sIgA concentrations (baseline = 52. On the other hand, other studies that included strength training found no significant differences [22,25]. Moreira et al. [25] only collected sIgA once a week, which affected the results, and Nunes et al. [22] did not increase the load over time unlike the study of Milanez et al. [23], which found some significant results.

Training and Official Matches
Some studies analyzed in-season periods that included training and matches [25,30]. He et al. [31] found that 4 weeks of intensive training period and 2 weeks with matches decreased sIgA when compared to recovery week concentrations (pre to post four weeks of training = 146.7 ± 18.0-144.9 ± 22.7; match 1 = 142.9 ± 11.9; match 2 = 153.2 ± 18.0; recovery weeks range values = 204.3 ± 20.5-210.7 ± 15.0 ug/mL). Another study found decreases in sIgA secretion (106 ± 20-92 ± 21 ng·mL −1 ), while absolute sIgA concentration increased (587 ± 94-720 ± 153 µg/mL) after 4 weeks of training with one match per week [26]. In this case, the increase in sIgA concentration could be associated with the lower load applied in the last week of training, which could have influenced results. If the authors collected saliva samples 1 week before, they probably would have found decreases in sIgA concentration [26].

Relationship between Salivary IgA and URTI
Few studies analyzed in the present systematic review found relationships between IgA and URTI [10,25,41]. However, these findings are inconsistent because other studies [28,29,37,38] found no statistically significant correlations. Furthermore, in general, the studies revealed methodological limitations in analyzing the relationship such as different times of salivary collection and different periods of analysis.
Moreover, it is important to reinforce that IgA is an antibody and an immune marker found on the mucosal surface, including saliva [46]. For that reason, IgA works as a protector against viruses and antigens [47]. Therefore, low levels of sIgA could be one of the reasons causing URTIs [10,25,41].
The study of Moreira, Moura et al. [25] analyzed a 4 week period of intensive training. Those authors hypothesized that increments in training load might decrease sIgA levels and increase symptoms associated with URTI, and they confirmed this hypothesis, finding such a correlation in week 4. This study indicated that some athletes could have decreased sIgA and, consequently, be exposed to higher risk of URTI. This argument was also stated by another study, which found that a low sIgA secretion rate may be considered a risk factor for the development of URTI [48]. In addition, it was found that higher load and intensity could also contribute to URTI occurrence [41,48]. The Tiernan et al. [10] study also corroborated this relationship, finding that a decrease ≥65% in sIgA contributed to a higher risk of developing URTI in the following 2 weeks.
In addition, Cunniffe et al. [28] did not present significant correlations, but the authors found lower sIgA concentrations in players who reported higher incidences of URTI than players who were URTI symptom-free.
Meanwhile, other studies could not present such results, which does not mean that a biological relationship does not exist [49]. Moreira et al. [37] analyzed a 17 day training period with respect to sIgA levels in basketball players, and only one episode of URTI symptoms was reported with no association with sIgA levels. Orysiak et al. [38] also analyzed the effects of a 17 day training period on sIgA levels in ice hockey players and found no statistically significant correlations between the URTI incidence and sIgA. The same study justified this by the variation in mucosal markers between players, which could have affected the results, and the authors reinforced an individual analysis, player by player. In addition, the small sample size may have also affected the results. Another study also confirmed these findings, in which lower concentrations of sIgA were observed in the athletes with URTI symptoms compared to the healthy state [29]. This helps to explain the previous suggestion to monitor mucosal markers at an individual level [11,13,29].
From the aforementioned studies, the hypothesis of the present study that IgA correlated with URTI is supported. It was possible to confirm that low levels of sIgA could be associated with a higher risk of URTI. In other words, it is recommended to avoid long periods with low levels of sIgA and, consequently, higher risk of URTI.

Relationship between Salivary IgA and Training Load
Another main finding of the present systematic review was that a higher training load was associated with low levels of salivary IgA, which confirmed our hypothesis. As suggested by Mariscal et al. [21], the determination of IgA concentrations could help to identify a higher or excessive training workload and could also determine the risk of URTI in professional athletes.
One of the aims of the Tiernan et al. [10] study was to investigate the relationship between sIgA and training load, and they hypothesized that there would be an inverse relationship between sIgA and training load. However, while there were no significant associations found between these markers, the same study showed that, 1-2 weeks before the decrease in sIgA, the training load increased by 49%. One justification for the lack of association between sIgA and training load could be associated with appropriate training load management, avoiding low levels of sIgA by ensuring sufficient recovery [50]. Furthermore, Lindsay et al. [20] found a relationship between sIgA secretion rate and player load. No explanations for this result were given by those authors. In addition, Mariscal et al. [21] did not find any relationship, but they found that participation for more than 30 min showed a statistically significant decrease in IgA. The authors suggested that cumulative activity could influence some biomarkers such as IgA; therefore, rotation of players should be considered for team sports to avoid this negative outcome.
The nonsignificant findings of previous studies [10,20,21] contradicted others [2,20,25,41]. For instance, Moreira, Moura et al. [25] found a significant association between training load and URTI in futsal athletes. As training load decreased, the URTI symptom severity decreased, suggesting that higher training periods led to athletes becoming more susceptible to developing an URTI. The previous statement was supported by other studies finding that higher load and intensity could also contribute to URTI occurrence [41,48]. The previous findings lead us to reinforce that higher levels of URTI are associated with low levels of salivary IgA, which are also associated with higher levels of training load.
Further information was given by Coad et al. [35], who revealed that lower values of sIgA occurred with higher player load values in rugby players. Moreira, Mortatti et al. [51] demonstrated that, compared with a post-season detraining phase, sIgA decreased when athletes were involved in preseason training and in-season training and competition. It is important to reinforce that this study analyzed a congested period (more than one match per week), which reinforces one of our study hypotheses. For instance, Morgans et al. [52] also revealed that compromised immunological function appeared during a highly congested soccer season. Despite soccer studies not being analyzed in the present systematic review, this study is mentioned because it supports the previous finding, necessitating more research in other sports. A less recent study found that 6 weeks of preseason training and a 10 week in-season schedule for American college football athletes significantly decreased sIgA and increased the incidence of URTI [53].

Study Limits and Future Directions
As mentioned by Nunes, Crewther, Ugrinowitsch et al. [33], there are some confounding effects of other variables such as type of training, periodization, and psychological factors that can influence results and should be considered for future research. However, except for He et al. [31], Koch et al. [34], and Yamauchi et al. [39], who used university or college athletes, and Moreira et al. [37], who included professional athletes and staff members, all studies analyzed in the systematic review represented the actual training environment of athletes and the inherent limitations.
In addition, there are other factors that can contribute to decreases in sIgA and, thus, increase the risk of infection, such as exposure to pathogens, mucosal damage, and environmental conditions (i.e., air pollutants and pollens). These factors could be involved in the higher incidence of respiratory infections [37].
However, the 24 studies included in this systematic review did not analyze studies on soccer due to the large number of published scientific manuscripts; therefore, all conclusions should be carefully interpreted by soccer researchers, coaches, or their staff members.
The present research only analyzed one study [28] that included a full season; thus, future studies should analyze full seasons in different sports.
Furthermore, the small sample size was a limitation identified in the studies included in the present systematic review, necessitating more research with a larger number of participants.
Moreover, future studies are required to clarify the adaptations in sIgA produced by different training protocols and types, as the present research could not highlight this information due to few studies clearly presenting the training protocol. Furthermore, more insight is required regarding congested periods (i.e., weeks with two or more matches) in order to clarify, expand, and confirm the present results, since no study in team sports other than soccer analyzed this variable.
The main confirmation of this study is that sIgA should be used as a measure to analyze the risk of a subsequent URTI in different sport teams. With that information, coaches, staff, and the scientific community could imply appropriate prescription and management of training load through a proper periodization. Thus, training and performance can be optimized, and players can be better prepared for competition. Saliva sample collection is also noninvasive, easy to collect, and time-efficient; thus, it could be easily implemented for training quantification [46]. Lastly, as suggested by Tiernan et al. [10], sIgA should be collected before the start of the season as a reference baseline for future collections and better interpretation of outcomes. Then, sIgA should preferably be collected before and after matches or in the beginning and ending of a microcycle. The results of sIgA and its relationships with URTI and training load should be individually analyzed for better interpretation.

Conclusions
The present systematic review confirmed both hypotheses, i.e., that lower values of IgA occurred after greater training load periods, and that IgA was correlated with URTI. In brief, IgA decrement is related with both greater intensity/volume and congested schedules. Therefore, physical fitness and conditioning coaches should carefully manage training load progression, avoiding high-intensity sessions on two consecutive days or following matches. In addition, since competition may suppose an additional stressor that may affect IgA, no high-intensity training sessions should be performed during at least the two days following competition.

Strength of the Study
The number of studies aimed at extracting hormone levels and/or antibodies (e.g., IgA) seems to be growing in popularity [4]. Antibodies are proteins that the immune system makes to fight attacks by bacteria, viruses, and toxins. The mucosal surfaces are protected by a network of organized structures located in the gut, urogenital tract, oral cavity, and respiratory system, collectively known as the mucosal immune system [14].
The production of secretory IgA has been considered as the "first line of defense" of the mucosal immune system against pathogens [54]. However, since different articles have highlighted a relationship between training/match load and IgA levels, and between lower levels of IgA and URTI, physical fitness and conditioning coaches should ensure an optimal training load management to avoid immunosuppression.
On this note, a recently published systematic review about recommendations and best practices to return to competition after COVID-19 lockdown briefly highlighted the importance of training load management to avoid effects on IgA [17]. However, to the authors' knowledge, despite the large number of articles published aimed at assessing the influence of training and competition load on sIgA, there is a lack of systematic reviews