Probiotic fermented milk improves immune function in marathon runners

Researchers have long recognized the immunomodulatory activity of certain probiotic strains (see review). A fermented milk that could improve immune parameters and reduce post-event symptoms would be a desirable dietary choice for endurance athletes.

In a recent human trial, Brazilian researchers set out to see if a probiotic fermented milk (Yakult) could improve immune responses in male marathon runners. Previously, Lactobacillus casei Shirota, the fermentative and probiotic strain in the fermented milk Yakult, had been shown to reduce the number of athletes who experienced one or more weeks with upper respiratory tract symptoms (Gleeson et al. 2011). Since the upper airway symptoms suffered by endurance athletes may be due to infectious, inflammatory and/or allergic conditions, Vaisberg and colleagues wanted to study how a probiotic might impact a cross-section of immune parameters in marathon runners.

The study parameters are as follows:

  • Intervention: Yakult fermented milk, containing 4×1010 cfu/d Lactobacillus casei Shirota consumed daily for 30 days compared to placebo, which contained no bacteria and was an unfermented milk indistinguishable to volunteers from the probiotic
  • Subjects: Fifty-six male volunteers, all of whom reported having exercise-induced upper respiratory tract symptoms, were included in the study. Forty-two (20 probiotic milk and 22 placebo) completed the trial. Baseline cardiorespiratory capacity, body composition analysis and most recent race time were recorded and used to stratify subjects into the two test groups.
  • Sampling times: Before intervention, after 30 days of intervention, immediately after race, 72 hr after race, 14 d after race.
  • Endpoints: Researchers followed both subjective daily reporting of respiratory and gastrointestinal symptoms and immune markers. The immune markers were measured from three sample sites: nasal mucosa and serum (9 cytokines were measured), and saliva [secretory IgA and antimicrobial peptides (lactoferrin, lysozyme, defensin-1, cathelicidin) were measured]

The study showed no differences in upper respiratory tract symptoms (even though fewer subjects in the probiotic group reported respiratory symptoms, comparison did not reach statistical significance), but the small number of study subjects likely precluded detection of any differences between the groups.

Some changes in immune responses were observed between the probiotic and placebo groups. Lower salivary secretory IgA and defensin-1 was observed in the placebo compared to probiotic immediately after marathon. Both these factors are helpful in the body’s ability to fight infection. The measurement of pro-inflammatory cytokines in nasal lavages revealed that IL-1B, IL-5, IL-6, IL-13, and TNF-alpha were lower and IL-10 (an anti-inflammatory cytokine) was higher immediately after the race in the placebo group compared to the probiotic group. The only difference between probiotic and placebo in serum cytokines was higher TNF-alpha (an inflammatory cytokine) in the placebo group. Taken together, these results suggest that the probiotic fermented milk was able to reduce inflammation in runners immediately after the race.

At several time points (after 30 days of probiotic consumption, immediately after the race and 72 hours after the race) higher neutrophil infiltration of nasal mucosa was observed in the placebo compared to probiotic group. Neutrophils are characteristic of acute inflammation.

This study provides mechanistic insight into how a probiotic fermented milk might be able to improve upper respiratory tract symptoms in endurance athletes through immune modulation. A larger study powered to show differences in clinical endpoints along with changes in immune parameters would be more convincing. Ultimately, it is improved symptoms (or improved performance) that would really motivate this group to consume a probiotic fermented milk prior to endurance events.

This study and written report had several additional limitations, which reduce confidence in the reported findings due to concerns about bias. This human trial was not registered (e.g., on, so it is not possible to verify a priori objectives with the published report. The published paper does not specify primary and secondary objectives, nor does it include a power calculation justifying the number of subjects recruited. Randomization and allocation concealment are hallmarks of a well-conducted clinical trial. But in this study, the study subjects were stratified according to several physiological variables and were allocated to the test or placebo groups based on this stratification. It does not appear that severity of exercise-induced upper respiratory tract symptoms were one of these factors. No reasons for study dropouts were given. Results were only presented in per protocol, not intention to treat. The proper conduct and reporting of human trials increases our trust in reported findings.