Probiotic Supplementation and Respiratory Infection and Immune Function in Athletes: Systematic Review and Meta-Analysis of Randomized Controlled Trials

Abstract

Objective

To evaluate the effectiveness of probiotic supplementation on upper respiratory tract infection (URTI) and inflammatory markers in elite athletes.

Data Sources

We searched the PubMed, EBSCOhost, Scopus, and Web of Science databases using the following terms: probiotic OR probiotics AND exercise OR sport OR athletes AND URTI OR respiratory infection OR URTIs OR inflammation OR inflammatory OR cytokines.

Study Selection

We screened the titles and abstracts of 2498 articles using our inclusion criteria. A total of 14 articles were selected for further analysis.

Data Extraction

For each study, 2 independent reviewers extracted the study design, participant characteristics, inclusion and exclusion criteria, intervention characteristics, outcome measures, and main results.

Data Synthesis

We did not observe an effect of probiotic supplementation on the number of days of illness or the mean number or duration of URTI episodes but did note an effect of probiotic supplementation on the total symptom severity score (−0.65; 95% CI = −1.05, −0.25; P = .02). Lower levels of interleukin 6 (−2.52 pg/mL; 95% CI = −4.39, −0.66 pg/mL; P = .002) and tumor necrosis factor α (−2.31 pg/mL; 95% CI = −4.12, −0.51 pg/mL; P = .01) were also reported after supplementation.

Conclusions

This meta-analysis provides evidence that probiotic supplementation, especially among professional athletes, is an effective way to decrease the total URTI symptom severity score. In addition, probiotic supplementation may decrease interleukin 6 and tumor necrosis factor α levels. More studies involving larger groups are needed to better assess this effect. The optimal timing, duration, composition, and dose of such supplementation need to be determined.

Key Points

• Probiotic supplementation, especially among professional athletes, effectively decreased the total symptom severity score of upper respiratory tract infection, especially when taken as a single-strain probiotic.

• Probiotic supplementation may decrease levels of interleukin 6 and tumor necrosis factor α.

• More studies with larger groups are needed to better evaluate this effect.

• Determination of the optimal timing, duration, composition, and dose of such supplementation is needed.

Among athletes, heavy, acute, and prolonged intense physical exercise; insufficient rest and sleep; emotional stress; and inadequate nutrition may generate serious health problems, such as inflammation or respiratory infections, that may affect their physical performance and sports achievements.¹ Researchers² have observed that such factors can lower the resistance of athletes by reducing the number and activity of natural killer cells, decreasing neutrophil activity, impairing proliferation of T lymphocytes, decreasing the level of anti-inflammatory cytokines, and increasing levels of proinflammatory cytokines and salivary immunoglobulin A (IgA), inducing respiratory tract infections. Indeed, many elite athletes reported substantial bouts of infections that interfered with their ability to compete and train.² Epidemiologic studies^3–5 also indicated that extensive training was associated with an increased risk of upper respiratory tract infection (URTI). Therefore, dietary strategies to improve the immune function of athletes and reduce their risk of URTI have been sought for many years.

Probiotics are live microorganisms that, when administered in adequate amounts, may confer a health benefit on the host.⁴ In a meta-analysis involving general adult populations, Hao et al⁵ showed that probiotics were better than a placebo in reducing the number of participants experiencing episodes of acute URTIs, the rate ratio of episodes of acute URTI, and antibiotic use. This suggests that probiotics may be more beneficial than a placebo for preventing acute URTIs.⁵ Wang et al⁶ also found that probiotic supplementation in children led to fewer days of URTI and fewer days absent from day care or school than did a placebo.

Furthermore, many investigators have demonstrated the beneficial effect of probiotic supplementation on inflammatory markers in the general population. For example, in a systematic review and meta-analysis, Kazemi et al⁷ showed that probiotics and synbiotics decreased some inflammatory markers. The intervention was most effective in reducing C-reactive protein (CRP) and tumor necrosis factor (TNF)-α in healthy or ill general populations.⁸ The use of probiotics in physically active individuals might serve as a strategy to further improve respiratory symptoms and inflammatory status and consequently enhance the physical performance of athletes.⁸ In a recent review, Sivamaruthi et al⁹ indicated that probiotic supplementation improved the immune system and reduced the severity and incidence of URTIs in athletes, but to the best of our knowledge, no authors have confirmed these results via a systematic literature review or meta-analysis. Therefore, the aim of our study was to evaluate the effectiveness of probiotic supplementation on respiratory infection and inflammatory markers in elite athletes based on data from randomized controlled trials and crossover studies.

METHODS

This systematic review was designed according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) recommendations and registered in the International Prospective Register of Systematic Reviews (No. CRD4202017669).

Search Strategy

We searched for relevant articles published through the end of September 2020 using the PubMed, EBSCOhost, Scopus, and Web of Science databases between February 1 and September 30, 2020. The databases were searched using the following key words and their various combinations: probiotic OR probiotics AND exercise OR sport OR athletes AND URTI OR respiratory infection OR upper respiratory tract infections OR inflammation OR inflammatory OR cytokines without restrictions. We also reviewed the reference lists of related reviews and original articles. The full search strategy is described in Figure 1.

Figure 1.

Study selection process.

Study Selection

The inclusion criteria were based on the Population, Intervention, Comparison, Outcome (PICO) framework.¹⁰ Studies were included if they (1) involved healthy adult professional athletes of both sexes (≥18 years); (2) incorporated interventions using probiotics; (3) involved a control or placebo group; (4) examined outcomes not previously defined (as an open question: all outcomes were reported in the studies); (5) focused on inflammatory variables, respiratory infections, or both; and (6) were randomized controlled trials (RCTs) and/or crossover studies.

Data Extraction

We independently performed a literature search and selection of articles. Publications were evaluated according to the titles, abstracts, and full texts in subsequent stages. Each selected publication was studied critically. During the data-abstraction process, we made no attempt to contact the authors for further information beyond what had been published. Discrepancies were resolved through consensus or arbitration. The data extracted from each study were as follows: authors, year of publication, cohort age and sex, exercise, intervention and control groups, main outcome, and results.

Bias Assessment

We used the Cochrane risk of bias assessment tool to judge the methodologic quality of each trial with the aim of evaluating the performance and methods of randomization, the extent of blinding (whether it affected data collectors, data analysis, outcome assessors, or participants), allocation concealment, incomplete outcome data, selective reporting, and other possible sources of bias. In line with the Cochrane handbook¹¹ criteria for judging bias risk, each study was classified as having a high, low, or unclear risk of bias.

Statistical Analysis

Statistical analysis was carried out using the Statistica (version 13.0; StatSoft) software. The therapeutic effect of probiotic supplementation on URTI and cytokine levels compared with the placebo was estimated using the standardized mean difference with a 95% CI. We examined publication bias through visual inspection of a funnel plot and then using the Egger test¹²; P < .05 indicated a significant publication bias. Heterogeneity across studies was measured using the Cochran Q statistic, with P < .10 implying a significant difference, and the I² statistic, with I² = 0% indicating no heterogeneity and I² = 100% indicating maximal heterogeneity. A random-effects model was selected when I² ≥ 50%. All statistical tests were 2 sided, and P values <.05 were considered statistically different.

RESULTS

Search Results

From the initial search strategy, we identified 2501 articles. A total of 14 RCTs or crossover trials met the inclusion criteria and were included in the final meta-analysis (Figure 1).

Population and Study Characteristics

Supplements in capsules^13–21 or beverages^19,22–25 were used in most studies. Sachets with probiotic bacteria were used in 1 study.²⁶ The number of study participants, their characteristics, duration of the intervention, type of supplement, and sport discipline are presented in Table 1. A total of 1309 study participants, 771 study participants from the supplemented group (PRO) and 538 from the placebo group (PLA), were involved in the 14 selected investigations. The mean age of participants ranged from 20.1 ± 1.5 years²⁰ to 37 ± 11 years.²⁶ Interventions were based on supplementation of probiotic bacteria, such as Bifidobacterium animalis,^16,26 B. bifidum,^13,16,17,25 B. lactis,¹⁷ B. longum,¹³ Lactobacillus acidophilus,^16,17,26 L. casei,^17,22 L. casei Shirota,^23,24 L. fermentum,^17,18 L. gasseri,¹³ L. helveticus,¹⁵ L. plantarum,^14,17 L. rhamnosus,^17,19 L. salivarius,²¹ Saccharomyces boulardii,¹⁷ Streptococcus thermophilus,¹⁷ and Bacillus subtilis.²⁰ Although the publication dates of the selected papers were unrestricted, all articles in this systematic review were published after 2007.

Table 1.

Summary of Studies of Probiotics for Upper Respiratory Tract Infection and Cytokines Concentrations in Athletes Compared With Control Participants (N = 1309) Continued on Next Page

Study, Type	Population	Intervention	Control	Main Outcome Measure	Intergroup Differences, Mean ± SD		Adverse Effect
					PRO	PLA
Gill et al22 (2016), crossover	Male runners (age = 26 ± 6 y)  PRO: n = 8  PLA: n = 8	Beverage containing 1 × 1011 Lactobacillus casei x 1 wk	Beverage without Lactobacillus casei	IgA, mg/L	527 ± 317	505 ± 170	a
				IgA secretion rate, μg/min	260 ± 183	264 ± 182
Gleeson et al23 (2011), parallel	Male and female runners, cyclists, swimmers, triathletes, racket sports, and team gamers  PRO: n = 32, age = 32 ± 14 y  PLA: n = 26, age = 25 ± 9 y	Drink containing 6.5 × 109 Lactobacillus casei Shirota x 16 wk	Drink without Lactobacillus casei Shirota	Episodes of URTI	1.2 ± 1.0	2.1 ± 1.2	a
				Total symptom severity score	49 ± 55	50 ± 32
				Duration of URTI, d	7.9 ± 5.5	7.6 ± 3.2
Gleeson et al21 (2012), parallel	Male and female runners, cyclists, swimmers, triathletes, racket sports, and team gamers  PRO: n = 27, age = 25 ± 5 y  PLA: n = 27, age = 24 ± 4 y	Capsules with 2 × 1010 Lactobacillus salivarius CFUs, 1.78 maltodextrin, and 0.01 g magnesium stearate x 16 wk	0.78 maltodextrin and 0.01 g magnesium stearate	Episodes of URTI	1.5 ± 2.3	1.4 ± 1.5	a
				8-wk IgA, mg/L	127 ± 101	143 ± 95
				8-wk IgA secretion rate, μg/min	63 ± 36	63 ± 41
				16-wk IgA	123 ± 62 mg/mL	128 ± 58 mg/L
				16-wk IgA secretion rate, μg/min	57 ± 29	59 ± 32
Gleeson et al24 (2016), parallel	Male and female runners, cyclists, swimmers, triathletes, and team gamers  PRO: n = 126 (58% men), age = 20.3 ± 0.2 y  PLA: n = 117 (59% men), age = 20.6 ± 0.2 y	Drink with 6.5 × 109 Lactobacillus casei Shirota x 20 wk	Drink without Lactobacillus casei Shirota	Episodes of URTI	0.7 ± 0.1	0.6 ± 0.1	a
				Total symptom severity score	62 ± 5	64 ± 6
				Duration of URTI symptoms, d	5.6 ± 0.4	5.9 ± 0.5
Haywood et al13 (2014), parallel	Elite rugby union players (age = 24.7 ± 3.6 y)  PRO: n = 30  PLA: n = 30	Capsules with 2.6 × 109 Lactobacillus gasseri CFUs, 0.2 × 109 Bifidobacterium bifidum organisms, and 0.2 × 109 Bifidobacterium longum organisms x 4 wk	Capsules without probiotic bacteria	Days of illness	3.4 ± 4.6	5.8 ± 6.6	a
				Total symptom severity score	8.7 ± 25.1	13.33 ± 14.6
				Duration of URTI symptoms, d	4.6 ± 3.1	7.2 ± 6.9
Huang et al14 (2019), parallel	Male triathlon teams  Study I   PRO: n = 9; age range = 19–24 y   PLA: n = 9; age range = 19–21 y	Capsules with 1.5 × 1010 Lactobacillus plantarum PS128 CFUs and 100 mg excipient of microcrystalline cellulose x 8 wk	400 mg excipient of microcrystalline cellulose	Study I, pg/mL			a
				IL-6	6.6 ± 0.7	9.3 ± 1.2
				IL-10	9.4 ± 1.9	8.7 ± 1.1
				TNF-α	10.5 ± 0.8	14.4 ± 1.8
	Study II   PRO: n = 8; age range, 19–21 y   PLA: n = 8; age range, 19–26 y			Study II, pg/mL
				IL-6	14.1 ± 1.3	19.7 ± 2.2
				IL-10	131.7 ± 14	84.6 ± 1.8
				TNF-α	15.2 ± 1.9	22.1 ± 2.4
Kekkonen et al19 (2007), parallel	Male and female marathon runners  PRO: n = 61, mean age = 40 y (range = 22–58 y)  PLA: n = 58, mean age = 40 y (range = 23–69 y)	3 × 108 Lactobacillus rhamnosus LGG CFU/milk-based fruit drink or 5 ×109 CFU/capsule x 12 wk	Milk-based fruit drink or capsule without probiotic bacteria	Days of illness	5.3 ± 6.8	3.9 ± 5.9	NR
				Episodes of URTI	0.7 ± 0.9	0.5 ± 0.7
				Duration of URTI symptoms, d	7.9 ± 7.1	6.3 ± 4.3
Khani et al25 (2018), parallel	Male sprint athletes (age = 21 ± 3 y)  PRO: n = 53  PLA: n = 3	2 × 109 Bifidobacterium bifidum (BIB2) CFUs x 12 wk	Fruit juices without probiotic bacteria	IgA, mg/L	253 ± 83	194 ± 42 mg/L	a
Michalickova et al15 (2016), parallel	Badmintonists, swimmers, cyclists, alpinists, athletes, karateists, kayakers, judoists, triathletes (men and women)  PRO: n = 20, age = 23.5 ± 2.7 y  PLA: n = 19, age = 22.8 ± 2.5 y	2 × 1010 Lactobacillus helveticus Lafti L10 CFUs x 14 wk	1% magnesium stearate and 99% maltodextrin	Episodes of URTI	4.92 ± 1.96	6.91 ± 1.22	NR
				Total symptom severity score	110.92 ± 96	129.73 ± 40.33
				Duration of URTI symptoms, d	7.25 ± 2.90	10.64 ± 4.67
				IL-10, pg/mL	365.2 ± 75.07	434.8 ± 71.51
Pugh et al16 (2019), parallel	Male and female runners  PRO: n = 12; age = 34.8 ± 6.9 y  PLA: n = 12; age = 36.1 ± 7.5 y	25 × 109 Lactobacillus acidophilus, Bifidobacterium bifidum, and Bifidobacterium animalis subsp lactis CFUs x 4 wk	Cornstarch	IL-6, pg/mL	10.96 ± 7.86	13.58 ± 12.9	a
				IL-10, pg/mL	5.27 ± 3.91	5.78 ± 3.26
Pumpa et al17 (2019), parallel	Elite rugby union athletes  PRO: n = 9; age = 27.0 ± 3.2 y  PLA: n = 10; age = 26.6 ± 2.9 y	60 × 109 viable bacteria Lactobacillus rhamnosus, L. casei, L. acidophilus, L. plantarum, L. fermentum, Bifidobacterium lactis, B. bifidum, and Streptococcus thermophilus and 250 mg of Saccharomyces boulardii x 17 wk	Microcrystalline iron oxide yellow, iron oxide red, and gelatin capsule and SB Floractivb (microcrystalline cellulose, lactose, calcium hydrogen phosphate dihydrate, povidone, silica colloidal anhydrous, magnesium stearate, and gelatin capsule)	Stage I: IgA, mg/L	487.2 ± 173.1	523.2 ± 192.3	NR
				Stage II: IgA, mg/L	452.4 ± 166.3	480.7 ± 200.6
Townsend et al20 (2018), parallel	Male baseball players (age = 20.1 ± 1.5 y)  PRO: n = 13  PLA: n = 12	1.2 × 109 Bacillus subtilis CFU/capsule x 12 wk	Maltodextrin	IL-10, mg/L	2.89 ± 1.08	3.27 ± 1.02	NR
				TNF-α, mg/L	2.07 ± 0.76	2.78 ± 0.95
				IgA secretion rate, μg/min	176.6 ± 86.6	156.1 ± 98.3
West et al18 (2011), parallel	Male cyclists and triathletes  PRO: n = 29; age = 35.2 ± 10.3 y  PLA: n = 33; age = 36.4 ± 8.9 y	1 × 1010 Lactobacillus fermentum (VRI-003 PCC) CFUs x 11 wk	Microcrystalline cellulose	Days of illness	0.37 ± 0.72	1.26 ± 1.9	a
				Episodes of URTI	0.6 ± 0.9	1.0 ± 1.5
				Total symptom severity score	1.23 ± 0.36	1.65 ± 0.57
				Duration of URTI symptoms, d	3.5 ± 6.6	7.4 ± 10.3
				IL-6, pg/mL	0.92 ± 2.14	1.22 ± 2.1
				IL-10, pg/mL	0.95 ± 1.69	1.16 ± 1.44
				TNF-α, pg/mL	1.27 ± 2.02	1.66 ± 2.22
	Female cyclists and triathletes  PRO: n= 18; age = 36.5 ± 8.6 y  PLA: n = 17; age = 35.6 ± 10.2 y			Days of illness	1.98 ± 2.63	1.08 ± 1.32
				Episodes of URTI	1.6 ± 2.1	0.5 ± 0.8
				Total symptom severity score	1.38 ± 0.45	2.07 ± 0.77
				Duration of URTI symptoms, d	12.2 ± 14.8	5.1 ± 14.7
				IL-6, pg/mL	0.71 ± 2.93	2.29 ± 4.8
				IL-10, pg/mL	0.89 ± 1.84	1.45 ± 2.03
				TNF-α, pg/mL	1.15 ± 2.0	1.71 ± 1.8
West et al26 (2014), parallel	Male and female athletes  PRO: n = 161; age = 36 ± 12 y  PRO1: n = 155; age = 36 ± 11 y  PLA: n = 149; age = 37 ± 11 y	PRO: 2 × 109 CFU of Bifidobacterium animalis subsp lactis Bl-04 PRO1: 1 × 1010 Lactobacillus acidophilus NCFM and Bifidobacterium animalis subsp lactis Bi-07 sucrose base CFU x 11 wk	Sucrose base without the probiotic bacteria	Days of illness	PRO: 4.4 ± 1.3 PRO1: 4.3 ± 1.9	4.1 ± 1.6	NR
				Duration of URTI symptoms, d	PRO: 6.1 ± 3.3 PRO1: 6.2 ± 4.8	5.7 ± 3.6

Abbreviations: CFU, colony-forming unit; IgA, immunoglobulin A; IL, interleukin; NR, not reported; PLA, placebo group; PRO, probiotic group; TNF, tumor necrosis factor; URTI; upper respiratory tract infection.

^aAdverse effects were not estimated.

^bFIT-BioCeuticals Ltd.

Effect of Probiotic Supplementation on Upper Respiratory Tract Infection

The 8 studies^{13,15,18,19,21,23,24,26} included in our meta-analysis measured the effects of probiotic supplementation on URTI. Data from the included articles allowed us to assess the effects of probiotic supplementation on the number of days of illness, the number of URTI episodes, total symptom severity score, and mean duration of URTI symptoms. The Begg rank correlation test (number of days of illness: P = .99; number of URTI episodes: P = .14; total symptom severity score: P = .99; mean duration of URTI symptoms: P = .29) and Egger linear regression test (number of days of illness: P = .66; number of URTI episodes: P = .31; total symptom severity score: P = .10; mean duration of URTI symptoms: P = .84) suggested that no publication bias was present.

In the groups receiving supplements, the days of illness, number of URTI episodes, total symptom severity score, and duration of URTI symptoms among supplemented groups ranged from 0.37 ± 0.72¹⁸ to 5.3 ± 6.8 days,¹⁹ from 0.6 ± 0.9¹⁸ to 4.92 ± 1.96,¹⁵ from 1.23 ± 0.36¹⁸ to 110.92 ± 96,¹⁵ and from 3.5 ± 6.6^15,18 to 12.2 ± 14.8 days,¹⁸ respectively. In the placebo group, respective ranges were from 1.08 ± 1.32¹⁸ to 5.8 ± 6.6 days¹³ for days of illness, from 0.5 ± 0.7¹⁹ to 6.91 ± 1.22¹⁵ for number of URTI episodes, from 1.65 ± 0.57¹⁸ to 129.73 ± 40.33¹⁵ for total symptom severity score, and from 5.1 ± 14.7¹⁸ to 10.64 ± 4.67¹⁵ days for duration of URTI symptoms.

We did not observe any effect of probiotic supplementation on the number of days of illness or on mean number or duration of URTI episodes, but we did note an effect of probiotic supplementation on total symptom severity score (Figure 2; Table 2). The effect remained after the analysis was performed on single-strain supplements only^15,18,23,24 (Figure 3; Table 2).

Figure 2.

Effects of probiotic supplementation on total symptom severity score in athletes with upper respiratory tract infections.

Table 2.

Effect of Probiotic Supplementation on Upper Respiratory Tract Infections

Variable	Meta-Analysis, Mean Difference (95% CI)	Value, %		P Value	No. of Studies
		I 2	Q
Total symptom severity score	−0.63 (−1.01, −0.26)	27.21	6.87	.02	5
Single strain	−0.64 (−1.04, −0.24)	36.18	6.27	.02	4
Days of illness	0.04 (−0.49, 0.58)	65.97	14.69	.88	4
Single-strain probiotic	0.25 (−0.17, 0.59)			.75	2
Multistrain probiotic	0.21 (−0.18, 0.60)			.30	2
Upper respiratory tract infection episodes (single strain only)	−0.21 (−0.61, 0.19)	82.48	34.25	.30	5
Duration of upper respiratory tract infection symptoms	−0.20 (−0.95, 0.56)	65.87	23.44	.61	7
Single strain	−0.51 (−2.04, 1.02)			.51	5
Multistrain	0.1 (−0.91, 1.11)			.85	2

Figure 3.

Effects of probiotic supplementation (single-strain probiotics only) on total symptom severity score in athletes with upper respiratory tract infections.

Effect of Probiotic Supplementation on Inflammatory Measures in Athletes

Nine studies^{14–18,20–22,25} in the meta-analysis quantified the effect of probiotic supplementation on inflammatory marker levels. Data from these articles allowed us to assess the effects of probiotic supplementation on interleukin (IL) 6, IL-10, TNF-α blood, and salivary IgA levels. The Egger linear regression test (IL-6: P = .94; IL-10: P = .06; TNF-α: P = .16; IgA: P = .74; salivary IgA: P = .40) suggested that there was no publication bias.

After intervention in the supplemented group, mean IL-6 ranged from 0.71 ± 2.93¹⁸ to 14.1 ± 1.3¹⁴ pg/mL, IL-10 from 0.89 ± 1.84¹⁸ to 365.2 ± 75.07¹⁵ pg/mL, TNF-α from 1.15 ± 2.0¹⁸ to 15.2 ± 1.9¹⁴ pg/mL, IgA from 123 ± 62²¹ to 527 ± 317²² mg/L, and salivary IgA from 57 ± 29²¹ to 260 ± 183²² μg/min. In the placebo group, mean ranged from 1.22 ± 2.1¹⁸ to 19.7 ± 2.2¹⁴ pg/mL for IL-6, from 1.16 ± 1.44¹⁸ to 434.8 ± 71.51¹⁵ pg/mL for IL-10, from 1.66 ± 2.22¹⁸ to 22.1 ± 2.4¹⁴ pg/mL for TNF-α, from 128 ± 58²¹ to 523.2 ± 192.3¹⁷ mg/L for IgA, and from 59 ± 32²¹ to 264 ± 182²² μg/min for salivary IgA (Table 1).

The meta-analysis did not show an effect of probiotic supplementation on IL-10 and IgA levels, but we found an effect of probiotic supplementation on IL-6 (I² = 81.99%, Q = 22.21, P = .002) and TNF-α (I² = 91.92%, Q = 49.49, P = .01; Figures 4 and 5, Table 3).

Figure 4.

The effect of probiotic supplementation on interleukin-6 (pg/mL).

Figure 5.

The effect of probiotic supplementation on tumor necrosis factor-α(pg/mL).

Table 3.

Effect of Probiotic Supplementation on Cytokine Levels

Variable	Meta-Analysis, Mean Difference (95% CI)	Value, %		P Value	No. of Studies
		I 2	Q
Tumor necrosis factor α, pg/mL	−2.31 (−4.12, −0.51)	91.92	49.49	.01	3
Interleukin-6, pg/mL	−2.52 (−4.39, −0.66)	81.99	22.21	.002	3
Interleukin-10, pg/mL	2.08 (−0.37, 4.52)	94.57	92.06	.10	4
Immunoglobulin A, mg/L	9.34 (−14.59, 33.28)	1.1	5.056	.44	4
Immunoglobulin A, μg/min	−0.6 (−13.15, 11.95)	0	0.356	.93	3

Bias Assessment

Most studies (93%) described the randomization process in detail and were judged as having a low risk of bias for this domain. Only 1 study demonstrated an unclear risk of bias.²⁵ Full details are presented in the Supplementary Table (see Supplemental Table, available online at http://dx.doi.org/10.4085/1062-6050-0592.20.S1).

DISCUSSION

Modulation of the immune system to increase defenses against URTIs is one of the most extensively researched areas in professional sports. We showed that probiotic supplementation positively affected both IL-6 and TNF-α levels. After probiotic supplementation, especially when single-strain probiotics were used, the total symptom severity score of respiratory infections was lower than in the placebo groups. However, we detected no effect of probiotic supplementation on the days of illness, number of URTI episodes, duration of URTI symptoms, or IL-10 and IgA levels (whether in blood or saliva).

In addition to the investigations cited in this meta-analysis, several other RCTs assessing the effects of probiotic supplementation on URTI and cytokine levels in elite athletes have been published. For example, Cox et al,²⁷ Komano et al,²⁸ and Strasser et al²⁹ reported that supplementation was associated with a reduction in the number of days, severity of respiratory illness, and incidence of URTI. Lamprecht et al³⁰ determined that probiotic supplementation beneficially affected TNF-α but not IL-6; this was in contrast to Jäger et al,³¹ who showed that probiotic supplementation resulted in an overall decrease in circulating IL-6. We excluded the IL-10 data of Michalickova et al¹⁵ from analysis because the mean value of this variable was extremely high. Shing³² identified no effects of probiotic supplementation on plasma concentrations of IL-6, IL-10, and TNF-α. However, given the lack of mean and SD values, we could not use these data in our meta-analysis.

Respiratory infections are known to substantially reduce the sport performance of athletes. We found that one of the basic reasons for the development of infections was weakening of the effectiveness of both specific and nonspecific humoral immune mechanisms related to the mucous membranes. Pyne and Gleeson³³ demonstrated that this manifested predominantly as a decrease in secreted salivary IgA, which has been linked to a higher incidence of URTI. In their 1-year observational study, Fahlman et al³⁴ suggested that a season of football training could result in decreases in both salivary IgA and the secretion rate of salivary IgA, as well as an increase in the incidence of URTI. Probiotic supplementation has received attention because of the suggestion that it can stimulate T-cell immune responses in vitro, which is important because many researchers^35–37 have shown that athletes have decreased numbers of T cells after intensive anaerobic exercise. In their review, Wosinska et al³⁵ emphasized that Bacteroides acidifaciens induced IgA production in murine models and, as a consequence, elevated the production of IgA+ B cells and B cells. These findings were significant because IgA plays a pivotal role in maintaining intestinal homeostasis, primarily by preventing the adherence of pathogens in the intestine. These results may suggest the mechanism of probiotic supplementation in the prevention of URTI. Unfortunately, our findings were the opposite of these reviews and reflected no increases in salivary or IgA blood levels after probiotic treatments compared with placebo. Given the limited number of publications, we were unable to perform subgroup analysis by sex, age group, design and duration of intervention or dose and type of probiotics; these factors may also have affected our findings.

Whereas recreational and moderate exercise may have anti-inflammatory and immunomodulatory effects, the intense exercise of elite athletes can induce inflammation through the synthesis and release of many cytokines (IL-6, IL-1β, macrophage inflammatory protein-1α, IL-8, TNF-α, IL-10, and IL-1 receptor antagonist).³⁸ In a recent meta-analysis that included adults with diabetes, Tabrizi et al³⁹ reported that probiotic supplementation lowered serum CRP and TNF-α and increased nitric oxide levels but did not affect IL-6 levels. However, in a meta-analysis, Milajerdi et al⁴⁰ showed that probiotic supplementation reduced serum concentrations of proinflammatory cytokines, including IL-6, IL-12, IL-4, high-sensitivity CRP, and TNF-α, but did not affect IL-1β, IL-8, interferon-γ, or IL-17 concentrations. In their meta-analysis, Nazari et al⁴¹ determined that probiotic consumption resulted in decreased plasma concentrations of IL-6 and TNF-α without an increase in interferon-γ (standardized mean difference = 0.43; 95% CI = 0.09, 0.76; P = .01). Milajerdi et al⁴⁰ also documented increased serum concentrations of IL-10 as an anti-inflammatory cytokine after probiotic supplementation. We observed similar results in our meta-analysis, that elite athletes had lower levels of TNF-α and IL-6 after probiotic supplementation; yet unlike Milajerdi et al,⁴⁰ we noted no changes in IL-10 concentration.

The beneficial effects of probiotic supplementation on IL-6 and TNF-α levels are particularly important for athletes, as an increase in IL-6 secretions is one of the causes of inflammation, fatigue, pain, mood changes, and concentration disorders.^31,42,43 Czarkowska-Paczek et al⁴⁴ demonstrated that physical exercise induced muscle damage and a complex cascade of nonspecific inflammatory responses. Interleukin-6 mRNA is detectable in skeletal muscle after prolonged, intense exercise, indicating that IL-6 is produced locally in the skeletal muscle.⁴⁴ The increases in plasma TNF-α concentration postexercise also suggest a source in damaged muscles.³⁵ The increased cytokine level may be associated with not only a decrease in the athlete's sport performance but also changes in immune health and an increased incidence of URTIs.³¹ Probiotics can help regulate inflammation in a number of ways, improving the structure and function of intestinal epithelial barriers.⁴¹ Probiotics and some of their secreted metabolic products can act as ligands for innate immune system receptors, directly affecting key proinflammatory pathways.⁴¹ Probiotic single-strain and multispecies probiotic supplementation activated the T and B lymphocytes and subsequently increased the production of important regulatory cytokines, including IL-10.^31,45,46

Our analysis also suggested that probiotic supplementation positively affected the total symptom severity score. The multistrain probiotic formulation could increase the chance of adhesion and colonization of the host by the probiotic strain. If the strains are compatible, they confer synergetic effects. However, in this case, the positive effect of probiotic supplementation on the total symptom severity score was primarily affected by single-strain supplements. We demonstrated no differences in the days of illness, the number of URTI episodes, or the duration of URTI symptoms.

To the best of our knowledge, this is the first meta-analysis to summarize the effects of oral probiotic supplementation on respiratory tract infection and inflammatory cytokines in elite athletes. Nazari et al⁴¹ attempted to assess the effects of probiotic consumption on inflammatory markers, but their research also included sedentary participants and nonprofessional athletes.^47,48 The strengths of our meta-analysis include the comprehensive literature search, the specified inclusion and exclusion criteria for the studies, the explicit methods of data extraction, the measures taken to reduce the influence of bias, and the assessment of heterogeneity.

Our work had certain limitations. First, as mentioned, few publications could be included in the meta-analysis. Second, there were many variations among the studies, including methods, duration of intervention, sport disciplines, type of training and follow-up, probiotic formulation, dose and duration of the intervention, formulation of probiotics, and functions of the strains. Some investigators gave supplements in the form of probiotic capsules, whereas others used beverages or probiotic sachets. Meybodi and Mortazavian⁴⁹ suggested that supplements were able to transfer large numbers of viable probiotics into the gastrodigestive tract without much loss during storage. In contrast, the viable number of probiotics in food products, especially fermented items, can considerably decrease after consumption, but foods provide their matrix protection to probiotics. Unfortunately, the best type of probiotic supplement is not yet known.

CONCLUSIONS

This meta-analysis has provided evidence that probiotic supplementation, especially among professional athletes, is an effective way to decrease the total symptom severity score of URTI, especially when this supplementation is taken as a single-strain probiotic. In addition, probiotic supplementation may decrease TNF-α and IL-6 levels. More studies with larger groups are needed to better estimate this effect. Moreover, the optimal timing, duration, composition, and dose of such supplementation need to be estimated.

SUPPLEMENTAL MATERIAL

Supplemental Table. Found at DOI: http://dx.doi.org/10.4085/1062-6050-0592.20.S1