Letter to the Editor: Methodological limitations in the RCT on synbiotics, immune parameters, and gut microbiota in healthy adults
This letter identifies statistical and sampling flaws in the original RCT that undermine conclusions about synbiotic-induced reduction of inflammatory markers.
| Population | Healthy adults enrolled in the original RCT by Li et al. (n=106) |
|---|---|
| Intervention | Formal methodological critique of the design, statistical analysis, and generalization of the RCT on synbiotic supplementation |
| Comparator | Established methodological standards (WHO BMI classification, GRADE, subgroup analysis, power calculation) |
| Outcome | C-reactive protein (CRP) — between-group difference; Interleukin-10 (IL-10) — between-group difference; Interferon-gamma (IFN-γ); Adequacy of BMI stratification; Statistical power and sample size calculation |
Summary of findings
| Outcome | Effect | 95% CI | Certainty | Clinical relevance | Notes |
|---|---|---|---|---|---|
| C-reactive protein (CRP) — between-group difference | p=0.058 to 0.088, not significant at alpha 0.05, in the IC 95% reported | — | Low | — | 1 studies |
| Interleukin-10 (IL-10) — between-group difference | placebo group -16% (p<0.001 within-group); between-group p=0.008; in the IC 95% reported; heterogeneity 75.7% across 22 prior trials | — | Low | — | 1 studies |
| Interferon-gamma (IFN-γ) | p value not consistently <0.05; inconsistent result noted by letter authors; in the IC 95% reported | — | Low | — | 1 studies |
| Adequacy of BMI stratification | non-WHO BMI cutoff used (<24 vs >=24); 4 WHO categories not applied | — | Very low | — | 1 studies |
| Statistical power and sample size calculation | not reported; n=106 without power justification | — | Very low | — | 1 studies |
Context
The original RCT (Li et al.) assessed synbiotics in healthy adults and concluded that CRP, IFN-γ were reduced and IL-10 was relatively increased. Methodological critique letters are formal instruments of scientific quality control. The issues raised concern inadequate stratification, non-significant p-values treated as significant, and absence of subgroup analyses.
What the study showed
The letter shows that the between-group difference for CRP did not reach p<0.05 (p=0.088 or 0.058), contradicting the authors' conclusion. For IL-10, the detected effect was relative: IL-10 remained stable in the synbiotic group while decreasing 16% in the placebo group (p<0.001 within placebo; between-group p=0.008), yielding a result dependent on the statistical model chosen. No power calculation was reported for n=106. Heterogeneity of 75.7% across 22 prior trials on IL-10 in healthy populations reduces generalizability.
How it was done
This is a methodological critique letter published in Gut Microbes (2023), with no primary data collection. It analyzes the double-blind RCT by Li et al. with n=106 healthy adults, 8 weeks of follow-up, stratified by sex and BMI (<24 or ≥24 kg/m²).
Effect magnitude
Not applicable: this document generates no independent effect size. The discussed effect (IL-10, between-group difference) has p=0.008 but no 95% CI is reported in the letter; the CRP effect does not reach conventional significance (p=0.058–0.088).
Risk of bias
The letter does not apply a formal risk of bias tool (RoB 2 not explicitly used). Critiques are valid but the letter does not reanalyze primary data, limited to pointing out inconsistencies. Table 1 inserted in the letter refers to variables from another study (schizophrenia, PANSS, HAMD), suggesting an editorial error that undermines the document's coherence.
What this study does NOT prove
This letter does not prove synbiotics are ineffective; it proves only that the original RCT does not support its stated conclusions on CRP and IL-10. No new primary data are presented.
In clinical practice
Clinicians should not adopt synbiotics for immune modulation in healthy adults based on this single RCT, given that primary outcomes (CRP) did not reach between-group significance. The relative decrease of IL-10 in the placebo group, interpreted as an increase in the intervention group, does not equate to a direct biological effect of the synbiotic.
Limitations
The letter does not apply a formal risk of bias tool (RoB 2 not explicitly used). Critiques are valid but the letter does not reanalyze primary data, limited to pointing out inconsistencies. Table 1 inserted in the letter refers to variables from another study (schizophrenia, PANSS, HAMD), suggesting an editorial error that undermines the document's coherence.
What is still missing
RCTs with greater statistical power, follow-up ≥12 weeks, WHO-compliant BMI stratification, and pre-specified subgroup analyses are required for robust conclusions.
Technical appendix
Version history
- 1.0 · 2026-06-24 — Auto-generated under Evidence Standard v1.0