Probiotics against candidiasis: systematic review of experimental and mechanistic evidence
Probiotics show consistent anti-Candida activity in predominantly in vitro models, but clinical evidence in humans is insufficient to support any practice recommendation.
| Population | Experimental models (predominantly in vitro; limited in vivo evidence) involving Candida spp., including C. albicans, non-albicans species, and C. auris |
|---|---|
| Intervention | Probiotics (Lactobacillus, Bifidobacterium, Saccharomyces) and postbiotics/probiotic-derived metabolites |
| Comparator | Varied experimental controls; no standardised clinical comparator |
| Outcome | Inhibition of fungal growth; Inhibition of biofilm formation; Inhibition of yeast-to-hypha transition; Inhibition of Candida adhesion; Host immune modulation; Activity against Candida auris and non-albicans species; Clinical efficacy in humans |
Summary of findings
| Outcome | Effect | 95% CI | Certainty | Clinical relevance | Notes |
|---|---|---|---|---|---|
| Inhibition of fungal growth | no pooled estimate available; qualitative consistent inhibition across in vitro studies | — | Very low | — | 18 studies |
| Inhibition of biofilm formation | no pooled estimate available; qualitative consistent inhibition in vitro | — | Very low | — | |
| Inhibition of yeast-to-hypha transition | no pooled estimate available; qualitative consistent inhibition in vitro | — | Very low | — | |
| Inhibition of Candida adhesion | no pooled estimate available; qualitative consistent inhibition in vitro | — | Very low | — | |
| Host immune modulation | no pooled estimate available; mechanistic data only, in the clinical immune endpoint quantified | — | Very low | — | |
| Activity against Candida auris and non-albicans species | emerging in vitro signals only; in the quantified pooled estimate; insufficient studies | — | Very low | — | |
| Clinical efficacy in humans | no clinical trials included; in the human efficacy data available | — | Very low | — |
Context
Candida spp. cause mucocutaneous infections and life-threatening invasive candidiasis with high mortality in immunocompromised hosts. Growing antifungal resistance and emergence of multidrug-resistant Candida auris drive demand for adjunctive strategies. Probiotics are under evaluation as microbiome-based options, but the clinical evidence base remains sparse.
What the study showed
Of 18 included studies, most used in vitro models. Probiotics — predominantly Lactobacillus spp. — consistently inhibited Candida growth, adhesion, morphological transition, and biofilm formation. Identified mechanisms include antimicrobial metabolite production, competitive exclusion, immunomodulation, and epithelial barrier reinforcement. Heterogeneity in strains, models, and outcomes precluded quantitative meta-analysis; no pooled estimates with 95% CI were calculated.
How it was done
Qualitative systematic review (PRISMA 2020) searching PubMed, Scopus, and Web of Science (January 2015 to January 2026). Eighteen studies met eligibility criteria and were included in qualitative synthesis only, with no meta-analysis performed. PROSPERO registration: CRD420261278363.
Effect magnitude
Meta-analysis was not feasible due to heterogeneity. No pooled effect size (RR, OR, SMD) with 95% CI was reported; effect direction is consistently favourable in experimental models, but clinical magnitude remains unknown.
Risk of bias
The review did not formally apply standardised risk-of-bias tools (RoB 2, ROBINS-I, or AMSTAR-2 for primary studies). The evidence base is predominantly preclinical (in vitro), with minimal in vivo evidence and no randomised clinical trials. High heterogeneity in strains, concentrations, models, and outcomes prevents generalisation and quantification.
What this study does NOT prove
This study does NOT prove clinical efficacy of probiotics in humans with candidiasis. In vitro results do not translate directly to clinical outcomes, and no specific strain can be recommended based on these data.
In clinical practice
There is insufficient clinical evidence to recommend probiotics as treatment or prophylaxis for human candidiasis. Clinicians must not substitute established antifungals with probiotics based on this review. Mechanistic evidence may inform the design of future clinical trials.
Limitations
The review did not formally apply standardised risk-of-bias tools (RoB 2, ROBINS-I, or AMSTAR-2 for primary studies). The evidence base is predominantly preclinical (in vitro), with minimal in vivo evidence and no randomised clinical trials. High heterogeneity in strains, concentrations, models, and outcomes prevents generalisation and quantification.
What is still missing
Randomised controlled trials with specific strains, standardised dosing, and clinically relevant populations (immunocompromised, antibiotic users) are required to establish clinical efficacy and safety.
Technical appendix
Version history
- 1.0 · 2026-10-10 — Auto-generated under Evidence Standard v1.0