Acute and Chronic Effects of Interval Training on the Immune System: A Systematic Review with Meta-Analysis

Simple Summary Interval training (IT) is a popular training strategy recognized by its positive effects on metabolic and cardiovascular system. However, there seems no consensus regarding the effects of IT on immune system parameters. Therefore, we aimed to summarize the evidence regarding the effects of IT on the immune system. As our many findings, an IT acutely promote a transitory change on immune cell count followed by reduced function. The magnitude of these changes seems to vary in accordance with IT type. On the other hand, the regular practice of IT might contribute to improve immune function without apparent change on immune cell count. Abstract Purpose: To summarize the evidence regarding the acute and chronic effects of interval training (IT) in the immune system through a systematic review with meta-analysis. Design: Systematic review with meta-analysis. Data source: English, Portuguese and Spanish languages search of the electronic databases Pubmed/Medline, Scopus, and SciELO. Eligibility criteria: Studies such as clinical trials, randomized cross-over trials and randomized clinical trials, investigating the acute and chronic effects of IT on the immune outcomes in humans. Results: Of the 175 studies retrieved, 35 were included in the qualitative analysis and 18 in a meta-analysis. Within-group analysis detected significant acute decrease after IT on immunoglobulin A (IgA) secretory rate (n = 115; MD = −15.46 µg·min−1; 95%CI, −28.3 to 2.66; p = 0.02), total leucocyte count increase (n = 137; MD = 2.58 × 103 µL−1; 95%CI, 1.79 to 3.38; p < 0.001), increase in lymphocyte count immediately after exercise (n = 125; MD = 1.3 × 103 µL−1; 95%CI, 0.86 to 1.75; p < 0.001), and decrease during recovery (30 to 180 min post-exercise) (n = 125; MD = −0.36 × 103 µL−1;−0.57 to −0.15; p < 0.001). No effect was detected on absolute IgA (n = 127; MD = 47.5 µg·mL−1; 95%CI, −10.6 to 105.6; p = 0.11). Overall, IT might acutely reduce leucocyte function. Regarding chronic effects IT improved immune function without change leucocyte count. Conclusion: IT might provide a transient disturbance on the immune system, followed by reduced immune function. However, regular IT performance induces favorable adaptations on immune function.


Introduction
It is widely accepted that moderate-intensity continuous training (MICT) with short to moderate duration (<60 min) is associated with an enhanced immune defense [1].
However, acute bouts of high-intense or high-volume aerobic exercise might provide transitory negative changes on immune cell count and function (lasting between 3 h to 72 h depending on the immune outcome) [2,3]. This might lead to immunosuppression and increased risk for infectious diseases [1,4,5].
The underlying mechanisms to exercise-induced immunosuppression, referred to as the "open window", are multifactorial and involve neuroendocrine and metabolic factors such as catecholamines, cortisol and growth hormones [3,6]. Immunosuppression usually occurs after intensive training protocols that result in increased levels of inflammation, metabolic and oxidative stress [4]. Therefore, it is important to study different aerobic training protocols since different physiological demands could have different impacts on immune function.
Interval training (IT) is an aerobic training strategy that usually consists in interspacing periods of high-intensity efforts with periods of rest or low-intensity exercise [7,8]. The rationality behind this strategy is to allow the accumulation of higher volume of vigorous exercise than those that could be achieved performing continuous exercise at high intensity [9]. Although current studies about the topic involve low-volume protocols [10][11][12][13], IT is usually performed near or at maximum individual's capacity, which might result in higher metabolic and hormonal stress in comparison with MICT [14].
During the past century, IT gained popularity in sports preparation [15]. This training strategy was widely adopted by coaches and athletes to train at workloads closer to their specific performance competition [15]. However, in recent decades the recommendations of IT performance have been extended to non-athlete's subjects as an effective strategy for health promotion [16,17]. Although compelling evidence from healthy and clinical populations have consistently shown that IT promotes metabolic and cardiovascular benefits in a similar or greater extent than MICT [10,[18][19][20][21], there seems to be no consensus regarding the effects of IT on the immunological system.
Previous studies suggest that IT might induce changes in immune function for a few hours after exercise cessation [22][23][24][25][26]. There is evidence of both positive [27,28] and negative [29,30] functional adaptation of the immune system in response to repeated IT sessions such as improvements on immune defense and reduced immune cell count or death. Considering these controversial findings, it remains to be elucidated how IT strategies might affect the immune system, especially considering the many different IT models [7]. This knowledge might contribute to optimize IT prescription in both health and disease, elucidating which aspects in IT prescription might impact on immune system modulation and help health professional to prescribe more efficient and safer exercise protocols to different populations. Therefore, we aimed to summarize the evidence through a systematic review of literature regarding the results of clinical trials that investigated the acute and chronic effects of IT on immune measures in humans. Additionally, a meta-analysis was conducted to determine the acute effects of IT on the relevant immune parameters that presented sufficient data.

Methods
The set of items of this systematic review are reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [31]. The study protocol was registered with the International Prospective Register of Systematic Review (PROSPERO; available at: https://www.crd.york.ac.uk/PROSPERO/, accessed on 5 July 2020) (registration number CRD42020176291) [32]. The study design followed PICO strategy: humans (Population), acute or chronic IT (Intervention), other exercise interventions, non-exercise control or without comparison group (Comparison), leucocyte count, neutrophil count or function, lymphocyte count or function (Outcomes).

Eligibility Criteria
Systematic search comprised studies such as clinical trials, randomized clinical trials and randomized cross-over trials in humans. Studies were considered eligible for inclusion according to the following criteria: (i) involved humans without restriction for age, sex or health condition (ii) involved at least one IT session (with comparison group or not), here defined as intermittent activities that interspersed maximal efforts (e.g., all-out sprints) or close to maximal efforts (≥80% of peak or maximal oxygen uptake ( . VO 2peak/max ) or ≥85% of peak or maximal heart rate (HR peak/max )) with passive or activity recovery [33]; (iii) appropriate measures of exercise intensity (i.e., heart hate, . VO 2 , performance markers, or rating of perceived exertion); (iv) investigated at least one outcome of acute or chronic IT intervention on innate or acquired immune system; (v) published in English, Spanish or Portuguese. Studies were excluded based on the following criteria: (i) clinical trial registers or non-concluded studies, dissertation and thesis, letter to editor, literature reviews and observational studies; (ii) involved concurrent (i.e., IT combined with resistance training) or polarized training (i.e., IT combined with MICT).

Search Strategy
English, Portuguese and Spanish language searches of the electronic databases Pubmed/Medline, Scopus, SciELO were initially performed in March 2020 with an update on November 2020. Articles were retrieved from electronic databases using key words and MeSH terms: ("high intensity interval training") OR "high intensity intermittent training") OR "high intensity interval exercise") OR "high intensity intermittent exercise") OR "sprint interval training") OR "repeated sprint training") OR "hiit") OR "interval training")) AND ("immune system") OR "neutrophil") OR "iga") OR "immunoglobulin") OR "macrophage") OR "monocyte") OR "leucocyte") OR "lymphocyte") OR "upper respiratory tract infection") OR "urti") OR "illness") OR "immunity"). To inception, retrieved articles on systematic search were checked for relevance by two independent researchers (DS and AFV). After excluding repeated references, articles were selected after a sequenced title and abstract reading, always in this order. The agreement rate between reviewers for the title/abstract screening was high (kappa = 0.944, p < 0.001). After, DS and AFV independently reviewed the full texts of potentially eligible papers, and a third researcher (PG) was consulted when there was any disagreement between reviews. Additionally, manual search was conducted through to the references of all included studies to obtain an integrative cross-references full-text selection.

Data Extraction
The following data were independently extracted by two authors (DS and AFV): study design, participants characteristics (age, sex, sample size, health status, level of physical conditioning), IT protocols description and outcomes of innate and acquired immune measures (leucocytes, neutrophils, lymphocytes, lymphocyte subsets, monocytes, eosinophil, basophil, granulocyte, immunoglobulin A (IgA)). Additionally, IT was classified in accordance with training characteristics. The IT protocols involving maximal sprints ("all out" effort) were classified as sprint interval training (SIT), while IT protocols involving submaximal efforts such as the intensities closer to those that elicit the maximum oxygen consumption (≥85% . VO 2max ) were classified as high-intensity interval training (HIIT). Studies were classified as acute and chronic interventions in accordance with their respective characteristics. Acute studies were defined as those that investigated the acute effects (usually transitory lasting up to 48h after exercise cessation) provided by a single IT session. While the chronic studies were defined as those that investigated the adaptations provided by accumulated IT sessions (at minimum 3 sessions). Parameters such as immune cell count referred to the quantity of immune cell, while immune cell function is associated with the immune response against stressor agents.

Study Quality
Study quality of randomized controlled trials that met inclusion criteria was independently assessed by two authors (AS and AFV) using the Tool for the Assessment of Study Quality and Reporting in Exercise (TESTEX) scale. The TESTEX scale is a validated tool (ICC ≥ 0.91, p < 0.001) specifically constructed for assessing the methodological quality of studies on physical exercise and training. This scale is composed by 15 points (5 points for studies quality and 10 points for methodologic reports) [34].
The scale considers this criteria's for punctuation: eligibility criteria specified (1point); randomization specified (1 point); allocation concealment (1 point); groups similar at baseline (1 point); blinding of assessor (1 point); measure at last one primary outcome in 85% subjects (until 3 points); intention to treat analyses (1 point); compare groups in at last one primary outcome (until 2 points); all outcomes are reported with points estimates (1 point); control patients are asked to report their levels of physical activity and their data are reported (1 point); exercise load is titrated to keep relative intensity constant (1 point); exercise volume and energy expenditure can be calculated (1 point) [34]. Thus, a 15-point maximum score can be obtained by each study. Additionally, we adopted a study quality classification adapted from previous systematic review with meta-analysis [21], where the punctuation obtained from each study was divided by 15 and subsequently multiplied by 100, resulting in a study quality expressed as percentage. Based on this, study quality was classified as low (<50%), fair (between 50% and 66.6%), and high (>66.6%). The study quality based on TESTEX punctuation was not used as exclusion criteria.

Statistical Analyses
A meta-analysis was conducted to determine the overall acute effects of IT on the immune outcome that presented a minimum of five trials, such as IgA concentration (µg·mL −1 ), IgA secretory rate (µg·min −1 ), total leucocyte and lymphocyte count (10 3 µL −1 ). The number of results regarding the chronic effects of IT on a specific immune outcome was not sufficient to perform a meta-analysis. The effects for meta-analysis were calculated using the pre-intervention to post-intervention mean change and were presented as mean difference (MD) and 95% confidence interval. When the pre-and post-intervention values were not reported, the study was excluded from the meta-analysis. If the values were available only in figure, the authors were contacted by email to data request. When the response was not provided, the numeric data was obtained from chart through data extraction software (GraphData 1.0, Brazil). To evaluate the biphasic characteristic of lymphocyte count (i.e., immediately increase followed by decrease), an additional effect was calculated pre-exercise to the first time point recovery immediately post-exercise (i.e., obtained between 30 and 180 min post-exercise). Random-effects model was preferred due the high methodological variation between studies. The meta-analysis between conditions (IT vs. non-exercise) was not performed due too few studies involved a non-exercise arm as control condition. Statistical heterogeneity of the treatment effect among studies was tested using the Chi-square test and the inconsistency I 2 test, in which p < 0.10 and values above 50% were considered indicative of substantial heterogeneity. A sensitivity analysis was conducted to determine the contribution of each study to the overall effect by successively removing de results of each study and using the data from the remaining studies. In addition, subgroup analyses were performed to detect the influence of participants sex, modality, or IT type. Analyses were conducted using the Review Manager software (RevMan 5.3, Nordic Cochrane, Denmark), and the accepted level of significance was (p < 0.05).

Included Studies
Initially, 174 records were retrieved through searches strategy. After removing duplicates, 172 articles were screened for title and/or abstract analyses. Within these, 130 studies did not meet inclusion criteria and were removed. Subsequently, two researchers (DS and AFV) independently reviewed full text of the 42 remaining studies, in which three studies were removed because involved polarized training [28,35,36], three studies lack appropriate IT protocol description [37][38][39] and one study involved cold water immersion [40]. As result, 35 studies were included in final qualitative analysis. From these, 18 studies were included in quantitative analysis, where 12 studies were clinical trials, and six studies were randomized cross-over trials. When the study involved more than one IT intervention (e.g., different IT protocol or separated by sex), the data obtained from each intervention was calculated as an independent trial in meta-analysis. All these steps are described in Figure 1.

Summary of Studies
Studies' characteristics are summarized in Tables 1 and 2. Twenty-three studies investigated exclusively the acute effects of IT, whereas 10 studies performed interventions lasting between 1 [29,41] and 26 weeks [42]. Two studies performed both acute and chronic investigations [23,27]. . VO 2max , maximal oxygen consumption; . VO 2 max, velocity associated to maximal oxygen consumption; V max , maximal velocity achieved during the incremental test. fMLP, formyl-leucyl-methionyl-phenylalanine. ↑ significant increase; ↓ significant decrease; ↔ no significant change.   . VO 2max , maximal oxygen consumption; v . VO 2 max, velocity associated to maximal oxygen consumption; V max , maximal velocity achieved during the incremental test. ROS, reactive oxygen species; ↑ significant increase; ↓ significant decrease; ↔ no significant change.

Salivary Immunoglobulin A
A qualitative description of the acute effects of IT on immune measures are presented in Table 1. Six studies verified no change on absolute salivary IgA concentration after IT [22,46,47,52,54,55], while three studies verified transitory increase lasting up to 30 min after exercise [23,24,43]. Regarding secretory rate of IgA, four studies verified no change [24,46,52,55], and two studies verified decrease after exercise [22,23]. Considering IT type, the acute decrease on IgA secretion rate was only observed after SIT [22,23], while no HIIT intervention reduced this parameter [46,52,55] (Figure 2B).

Leucocyte Count
Ten studies verified transitory increases in total leucocyte count lasting up to 6 h after SIT [44,45,48] or HIIT [27,50,51,56,58,59,61]. One study verified no change on leucocyte count after a HIIT protocol with passive or active recovery [44]. Additionally, Fry et al. [61] reported a significant increase on leucocyte count immediately after HIIT when the highintensity bouts were performed at 120% of V max , but not at 90% ( Figure 2C).

Leucocyte Function
Five studies involving HIIT [26,27,57,60,61] reported a transitory reduction in lymphocyte function or reduced cell viability after IT performance (lasting up to 3 h) in response to in-vitro stimulation. One study found mobilization of low differentiated T cells and regulatory T cells (Treg) immediately after HIIT, in parallel with apoptosis of high differentiated T cells three hours after exercise [49]. Two studies verified transitory reduced neutrophil function after SIT [24] and HIIT [25] performance (lasting up to 30 min) in response to in-vitro stimulation (Table 1).

Qualitative Analysis of Chronic Effects of IT on Immune Outcomes
A qualitative description of the chronic adaptations on immune measures in response to IT is presented in Table 2. Two studies involving SIT reported no change on salivary IgA (absolute concentration or secretory rate) after training [23,63]. Three studies involving HIIT found no significant change in leucocyte count [30,42,65]. Regarding leucocyte function, one study verified increases on peripheral lymphocyte T helper subsets (i.e., memory regulatory T cell and Treg) [41] after HIIT. Three studies involving HIIT provided significant improvements on neutrophil function [62,64,66] and two studies involving HIIT [27,67] verified improvements on lymphocyte function. In contrast, a study involving three consecutive days of HIIT performed until exhaustion reported a significant increase on lymphocyte migration and apoptosis after the third day of consecutive training session [29].

Quality Assessment
Considering the specificity of the TESTEX scale, only 14 studies were included in this analysis and the results are shown in the Table 3. The studies achieved an average score of 4.6 from a total of 15 points. Point estimate of outcomes and exercise volume were the most reported features in the included studies. Most studies failed to report if there were, or not, adverse events associated with exercise intervention or intention to treat analysis. Table 3. Study quality and reporting of randomized clinical trial included studies.

Sensitivity Analysis
After sensitive analysis performance that checked outlies studies by successively removing the results of each study, changes were observed in effects of IT on absolute IgA concentration (p-value ranged from 0.001 to 0.14) and IgA secretory rate (p-value ranged <0.001 to 0.1) but not on total leucocyte count (p < 0.001), lymphocyte count immediately after exercise (p < 0.001), and during recovery (p-value ranged from <0.001 to 0.004).

Discussion
The aim of the present study was to summarize the evidence through systematic review of the experimental studies that investigated the acute and chronic effects of IT on immune measures. The main findings regarding acute studies were: (i) IT compromises IgA secretory rate but not IgA absolute concentrations; (ii) IT promotes transitory leukocytosis (lasting up to 6 h); (iii) IT promotes lymphocytosis followed by transitory lymphopenia (lasting up to 6 h); (iv) IT promotes a transitory impairment on lymphocyte and neutrophil function. Regarding chronic studies: (i) there are no changes on mucosal immune measures (IgA secretory and concentration) after repeated IT sessions spanning from 8 to 12 weeks; (ii) performing IT for 1 to 24 weeks provide no change on leucocyte count; (iii) chronic IT performance promotes favorable adaptations on lymphocytes, monocytes, and neutrophils function.
Both salivary secretory IgA and salivary IgA concentration play a major role in mucosal immune system and their levels have been inversely associated with occurrence of upper respiratory tract infection (URTI) [68,69]. The assumption that a single session of IT could compromise the salivary IgA due its high-intensity nature was supported by the meta-analysis; however, this seems be true only for IgA secretory rate ( Figure 2B). Whereas meta-analysis showed significant depression on IgA secretory rate immediately after exercise, there was no change on absolute IgA concentration (Figure 2A). The levels of absolute IgA concentration might even increase after a HIIT session ( Table 4).
The substantial heterogeneity observed in the results regarding the acute effect of IT on salivary IgA would be associated with large variety of methodological aspects of studies analyzed. For example, our subgroup meta-analysis revealed a different response of salivary IgA between sex after IT with a significant effect on IgA secretory rate for women but not for men (Table 4). This finding reinforces the role of sex on mucosal immunity modulation [52], which may be associated with differences on hormonal and/or autonomic nervous system activity between sexes [70,71]. Moreover, subgroup analysis detected significant effect on IgA secretory rate for SIT but not for HIIT, which suggest that the different IT types impact differently in this parameter (Table 4). Other methodological issues might also contribute to different results between studies such as dehydration, saliva method collection, or how IgA is expressed, as previously stated [4]. These findings should be interpreted with caution since the sensitive analysis detected the presence of outlier studies.
It is important to note that the acute impairments in salivary secretory IgA rate verified after a single SIT session has not been associated with occurrence of URTI [22,23]. This suggests that the magnitude of the observed transitory depression in salivary secretory IgA after SIT has no clinical relevance. A similar result was also confirmed after 8 weeks of SIT [23], showing that SIT could be performed three times per week in alternated days without altered susceptibility for URTI. Whereas there are some controversial results regarding the acute effects of SIT on salivary IgA secretory rate [22][23][24]52], HIIT seems to not impair IgA secretion rate or absolute IgA concentration in both trained and sedentary populations [43,46,55].
There is evidence that regular practice of IT might confer health and performance enhancing effect (e.g., aerobic and anaerobic capacity) without compromise the mucosal immune system in both athlete [23] and non-athlete population [63] (Table 2). Despite the study by Born et al. [28] involving polarized training and not meeting criteria to be included in this systematic review, it reveals interesting findings regarding the positive adaptation of the mucosal immune function in trained runners after nine HIIT sessions. Considering that the addition of IT into habitual aerobic training routine improved mucosal immune resilience to stress in parallel with the improvements on . VO 2max [28], it is reasonable to suggest a relationship between changes in cardiorespiratory fitness and modulation of immune mucosal function.
The leukocytosis observed after a single IT session was supported by our quantitative analysis ( Figure 3A). The early and rapid increase in blood leucocyte count after IT session might result from lymphocytosis, as well as detachments of neutrophil and monocytes from blood vessels caused by a high shear stress and catecholamines production, while the prolonged late increase seems to be induced by increases on cortisol levels that release neutrophils from bone marrow [3,4]. However, these responses might differ in magnitude, time course or duration depending on IT type [44], as well as body composition and cardiorespiratory fitness of the participants [51] (Figure 2C).
Whereas neutrophil count increases after an IT session, neutrophil function is transiently reduced in response to in-vitro stimulation [24,25] (Table 1). This functional impairment may be partially explained by the increased release of functionally immature neutrophil from bone marrow or by direct mechanisms such as stress hormones and oxidative stress [24]. However, it is not clear if this transitory functional impairment is clinically relevant. Moreover, neutrophil function is completely restored 24 h after exercise [25]. In contrast to acute findings, chronic IT effects might result in beneficial adaptations on neutrophil function such as improved stimulated reactive oxygen species (ROS) production, improved chemotaxis and reduced basal ROS production in both young and aging people [62,64,66] (Table 2). Recently, Bartlett et al. [62] verified improvements on neutrophil function after 10 weeks of a low-volume IT protocol in people with pre-diabetes.
A common concern regarding intensive aerobic training is its effects on cell-mediated immunity [17]. Although this is still a matter of debate [72], reduced lymphocyte count and function are usually associated with immunosuppression and increased risk for illness [1]. The present meta-analysis verified significant decrease in lymphocyte count during IT recovery (lymphopenia) ( Figure 5C). On the other hand, subgroup analysis suggests that HIIT might not necessarily promote lymphopenia, despite its high intensity characteristic (Table 4). Even in absence of lymphopenia, lymphocytes may become more vulnerable to stressful agents few hours after an IT session [26,27,57,60] (Table 1). These transitory functional impairments might be partially explained by changes in lymphocyte subset (e.g., reduced T-lymphocyte CD4+ and increased natural killer cells), with an impaired response to specific antigens [57,60]. Direct mechanisms such as lymphocyte redox imbalance [26] and stress hormone production [49] might contribute to impaired lymphocyte function during IT recovery. However, lymphocytes might adapt to repeated IT sessions and become more resistant to stress [27,42]. Moreover, impaired lymphocyte function is transient and returns to basal levels a few hours after exercise cessation [26,27,57,60]. Apoptosis of high differentiated lymphocyte T-cells after an acute IT session concomitantly with increased Treg cell count and progenitor cells suggest that IT could acutely impair immune response against latent infection, while improving immune defense against new invading infectious agents [49].
IT performed in both alternate [27,42] and consecutive days [67] might improve lymphocyte function. In contrast, IT performed until exhaustion in consecutive days may impair immune restoration and exacerbate lymphocyte migration and apoptosis [29]. These controversial results might be explained by diversity of IT protocols [7], which might result in different physiological responses [73][74][75]. Of note, studies reporting chronic improvements in immune system have reported no changes or reduction in the levels of stress hormones (e.g., cortisol and catecholamines) over long term [42,67]. This suggests that there was appropriate interval rest between IT sessions, since insufficient recovery is associated with chronic increases of these hormone levels at rest [76]. The adequate recovery might be dependent of IT type, protocols that promote higher increases in oxidative stress and stress hormone response may require more time to immune system restoration in comparison with less stressful IT protocols.
In short term, IT contribute to increase the Treg frequency in individuals with impaired metabolic profile (e.g., men with obesity) [41]. These findings are particularly important since Treg plays a key role on immune function regulation and its low levels are associated with impaired immune response [77]. Although IT may acutely have a negative impact on immune functioning, the increased susceptibility for illness seems be more associated with training schedule than an IT session per se. Considering its physiological demand, high training frequency or insufficient recovery between IT session might contribute to increased illness risk [29], while proper IT prescription might provide increased physical performance concomitantly with improvements or preservation of immune system [28,78].
As a practical recommendation, IT protocols that promotes elevated metabolic stress (e.g., high levels of cortisol, lactate, and sympathetic activation) should be avoided when it is desirable to preserve immune function such as in patients with depressed immune function or at imminent infection risk. In this sense, IT protocols involving short bouts (≤60 s) at submaximal efforts (≤90% of parameter associated with ( . VO 2max ) and total sessions with duration no longer than 60 min seems to be recommended. Whereas IT sessions involving "all out" efforts (≥30 s) seem to promote a greater disturbance on immune system.
There is compelling evidence that cardiorespiratory fitness is closely associated or can modulate immune functioning [28,41,[79][80][81][82][83], such that improved cardiorespiratory fitness might decrease the risk of illness. Therefore, a proper IT prescription might provide a time-efficient strategy to increase cardiorespiratory fitness while preserve or improve immunological function. In this sense, the detailed description of the effects of different IT types on several immunological parameters might contribute to provide valuable findings regarding proper IT prescription in immune system context. Whereas some acute parameters change after IT seem not clinically relevant per se, to understand their behavior should contribute to the maintenance of a sustainable exercise routine over the medium and long term.
This systematic review with meta-analysis was not free from limitations. Inclusion criteria resulted in heterogeneous studies, and conclusions could not be made to a specific effect of IT on immune system. In this sense, the variety of IT protocols, study designs, and outcomes might compromise the external validity of our analysis. Secondly, most studies included in this systematic review with meta-analysis have low quality and used relatively small sample sizes. Future studies should improve their methodological quality to provide reliable conclusions regarding the effects of IT on immunity. While further research is warranted to investigate the association between IT and illness risk. On the other hand, to the author's knowledge, this is the first study to summarize the state-ofthe-art knowledge available currently, regarding the effect of IT on immune system, which might bring relevant contributions to research area and clinical practice.

Conclusions
Based on our systematic review with meta-analysis of available literature, a single session of IT might provide a transient disturbance on the immune system, followed by reduced immune function. On the other hand, regular IT performance induces favorable adaptations on immune function, improving immunosurveillance in the short to long term without changing immune cell count.