Twenty four elite male combat athletes were randomized double blind to a seven strain probiotic blend at 1.5 x 10^10 CFU twice daily or placebo for four weeks spanning pre competition weight loss.
GI symptom scores diverged sharply between groups (GSRS total score group by time interaction F=28.829, p<0.001), driven largely by the placebo group deteriorating while the probiotic group roughly held steady.
Zonulin (p<0.001), D-lactate (p<0.001), LPS (p=0.001), TNF-alpha (p=0.005) and IL-6 (p=0.001) all moved favorably in the probiotic arm while secretory IgA rose (p<0.001), and butyrate producing genera were enriched. The strains were L. rhamnosus R0011, L. helveticus Lafti L10, L. salivarius HA-118, L. plantarum Lp-G18, L. helveticus R0052, B. animalis Lafti B94 and B. coagulans GBI-306086. Funding was the Shanghai Science and Technology Commission Project 23010504400 and the Shanghai Key Laboratory of Human Performance, with no commercial relationships declared.
The numbers
| Intervention | seven strains at 1.5 x 10^10 CFU twice daily for four weeks |
| GSRS total score, group by time interaction | F=28.829, p<0.001 |
| Zonulin | p<0.001 |
| Secretory IgA | p<0.001 |
| D-lactate | p<0.001 |
| LPS | p=0.001 |
| IL-6 | p=0.001 |
| TNF-alpha | p=0.005 |
What this does not show
- That anyone performed better. These are biomarkers plus a symptom score. No performance outcome was measured.
- That probiotics improve a normally fueled gut. The placebo group got substantially worse during the weight cut while the probiotic group roughly held steady. That is protection under an extreme stressor, not improvement in a normally fueled athlete.
- That the barrier was fully protected. Post intervention zonulin remained above the healthy reference range even in the probiotic arm, so held together is relative, not absolute.
- That the microbiome shifts drove the outcomes. Microbiota to outcome correlations did not survive false discovery rate correction, which is the appropriate standard. The butyrate producer enrichment is suggestive only.
- That zonulin measures what its name implies. Zonulin is a contested marker with known assay specificity problems, and the same caution applies in weaker form to D-lactate and secretory IgA.
- Anything about other populations or interventions. n=24, males only, one sport, four weeks. This is a small mechanistic probiotic trial; the result does not transfer to fibers, other supplements, or other athlete groups.
Caveats worth holding
- Small mechanistic trial: n=24, males only, one sport, four weeks.
- The effect is protection under the extreme stressor of a weight cut, with the placebo group deteriorating, not improvement in normally fueled athletes.
- Post intervention zonulin remained above the healthy reference range even in the probiotic arm.
- Microbiota to outcome correlations did not survive false discovery rate correction.
- Zonulin, D-lactate and secretory IgA are all imperfect markers, zonulin especially, with contested assay specificity.
- No performance outcome was measured.
- The study context is competitive weight cutting, a practice with real health risks; the findings describe biology under that stressor and are not an endorsement of the practice.