Highlights
– Updated Cochrane review (2025) of 47 randomized trials (n = 15,260) suggests probiotics may reduce C. difficile-associated diarrhea (CDAD) incidence with a relative risk 0.50 (95% CI 0.38–0.64) but low-certainty evidence.
– Absolute risk reduction (ARR) is small (1.6%); number needed to treat (NNT) ≈ 65 (95% CI 48–97). Benefits vary by baseline CDAD risk.
– Probiotics probably do not increase adverse events and may modestly reduce antibiotic-associated diarrhea (AAD); evidence limited by risk of bias, selective reporting, and industry ties in many trials.
Background and clinical context
Clostridioides difficile is the leading cause of healthcare-associated infectious diarrhea and can range from mild diarrhea to fulminant colitis and death. Antibiotic exposure is the dominant modifiable risk factor because it perturbs the gut microbiota and reduces colonization resistance. Preventing C. difficile-associated diarrhea (CDAD) during or after antibiotic use is therefore a major clinical goal across inpatient and outpatient settings.
Probiotics—live microorganisms that, when administered in adequate amounts, confer a health benefit—have been proposed as a strategy to preserve or restore gut microbiome resilience during antibiotic therapy. Multiple probiotic species and preparations (Lactobacillus spp., Saccharomyces boulardii, multi-strain combinations) have been studied, but trials vary in strain, dose, timing relative to antibiotics, patient population, and outcome definitions.
Study design of the evidence update
This update (Esmaeilinezhad et al., Cochrane Database Syst Rev 2025) searched CENTRAL, MEDLINE, and Embase through 3 March 2025 and included randomized controlled trials in adults and children receiving systemic antibiotics for any indication. Forty-seven trials (n = 15,260) met inclusion criteria; eight were newly identified since the previous review. Comparators included placebo, no treatment, or alternate prophylaxis.
Critical and important outcomes: incidence of CDAD (primary), C. difficile colonization, adverse events, antibiotic-associated diarrhea (AAD), and hospital length of stay. Risk of bias was assessed with Cochrane RoB 1; evidence certainty used GRADE. Data were pooled using random-effects models; sensitivity analyses explored missing data and heterogeneity.
Key results
Primary outcome: CDAD incidence
Pooled data from 38 trials (13,179 participants) found a relative risk (RR) of 0.50 (95% CI 0.38–0.64; P < 0.001) for CDAD with probiotic versus control. This corresponds to an absolute risk reduction (ARR) of 1.6% (probiotic group 1.6% [110/6787] vs control 3.2% [203/6392]). The number needed to treat (NNT) to prevent one case of CDAD was 65 (95% CI 48–97).
Certainty: Low. Downgrades were applied chiefly for imprecision due to a low number of CDAD events and for potential publication bias. About 27 of the 38 trials used for CDAD outcomes had missing outcome data (2%–45%), although sensitivity analyses that modeled plausible missing-data scenarios did not materially change the effect estimates.
Secondary outcomes: colonization, AAD, adverse events, and length of stay
C. difficile colonization (detection in stool without symptoms): 16 trials (1,302 participants) suggested a modest, non–statistically significant reduction (RR 0.87, 95% CI 0.68–1.11; ARR 2.1%; P = 0.27). Certainty: Low.
Antibiotic-associated diarrhea (AAD): 40 trials (13,419 participants) showed a larger relative effect (RR 0.67, 95% CI 0.57–0.78; P < 0.001) and an ARR of 9% (probiotic 23% vs control 27.4%). Certainty: Low, downgraded for heterogeneity and possible publication bias.
Adverse events: 37 studies (11,911 participants) reported adverse events. Pooled results suggested probiotics probably reduce adverse-event incidence slightly (RR 0.86, 95% CI 0.72–1.01; P = 0.074), absolute reduction 1.7% (moderate-certainty evidence). Reported events were generally mild (abdominal cramps, flatulence, nausea, soft stools, taste disturbance); serious probiotic-related events were rare in enrolled populations, which largely excluded immunocompromised patients.
Length of hospital stay: Seven trials (6,553 participants) found little to no difference (mean difference −0.07 days, 95% CI −0.35 to 0.21; moderate-certainty evidence).
Subgroup analyses, heterogeneity, and applicability
Investigators conducted a priori subgroup analyses by probiotic dose, species, adults versus children, inpatient versus outpatient, trial risk of bias, and baseline control-group CDAD event rate (low 0–2%, moderate 3–5%, high >5%). There was variability in baseline control event rates across trials (<2% to >5%). Because the absolute benefit depends on baseline risk, the clinical impact will be greater in higher-risk settings (higher baseline CDAD incidence) and modest where baseline risk is low.
Heterogeneity was present across trials in probiotic preparations, dosing regimens, timing relative to antibiotic start, outcome definitions and diagnostic methods for CDAD, and follow-up durations. Many trials were small, and nearly one-third lacked a published protocol or registration; 28 trials disclosed author affiliations or funding from probiotic companies, raising concerns for bias.
Safety considerations
Across trials enrolling non–immunocompromised adults and children, serious adverse events attributable to probiotics were uncommon. Typical gastrointestinal adverse effects were similar or slightly less frequent with probiotics. However, safety data remain limited for high-risk populations: severely immunocompromised patients, those with central venous catheters, short-bowel syndrome, or critical illness—groups in which case reports and small series have documented rare events such as fungemia (reported with Saccharomyces boulardii) or bacteremia with probiotic strains.
Expert commentary and interpretation
Mechanistic plausibility supports the observed clinical signal: probiotics may help maintain niche competition, produce antimicrobial metabolites, modulate local immune responses, and restore short-chain fatty acid production, all of which can strengthen colonization resistance against C. difficile after antibiotics. However, mechanistic effects are strain-specific and depend on timing and dose; therefore, pooling across heterogeneous probiotic products complicates inference about optimal agents.
The magnitude of benefit is modest on an absolute scale in many settings. With an NNT of ≈65, routine prophylactic use in low-CDAD-risk populations is likely to prevent few cases while adding cost and pill burden. In higher-risk settings (e.g., wards with elevated CDAD incidence, during high-risk antibiotic exposures such as clindamycin or fluoroquinolones, or in older hospitalized patients), the absolute benefit could be clinically meaningful.
Given limitations—low certainty for the primary outcome, variability in probiotic formulations, selective reporting in some trials, and industry involvement in many studies—recommendations should be cautious. Clinicians should weigh baseline CDAD risk, patient vulnerability, and probiotic strain evidence when considering prophylaxis. Importantly, routine use in immunocompromised or critically ill patients cannot be endorsed without larger safety data specific to those populations.
Limitations of the evidence and research gaps
Key limitations include: low event rates leading to imprecision; heterogeneous probiotic interventions preventing definitive guidance on optimal strains/doses; missing outcome data in many trials; potential publication and sponsorship bias; and underrepresentation of high-risk or immunocompromised cohorts. Few trials were large, placebo-controlled studies powered for CDAD as the primary endpoint.
Priority research needs: large, adequately powered, placebo-controlled trials focused on clinically meaningful endpoints (CDAD incidence, severe outcomes), stratified by baseline risk and using standardized diagnostic criteria; head-to-head comparisons of specific probiotic strains and dosing strategies; safety trials in higher-risk groups; and mechanistic microbiome studies linking ecological changes to clinical protection.
Conclusion and clinical takeaways
The 2025 Cochrane update indicates probiotics may halve the relative risk of CDAD in patients receiving systemic antibiotics, but the absolute benefit is small (ARR 1.6%; NNT ≈65) and evidence is low certainty. Probiotics probably do not increase short-term adverse events in non–immunocompromised populations and may reduce general antibiotic-associated diarrhea.
Clinical application should be individualized: consider probiotic prophylaxis when baseline CDAD risk is appreciable and when using strains and doses with supportive trial data; avoid routine prophylaxis in low-risk patients. Avoid probiotic use in severely immunocompromised persons or those with other risk factors for probiotic-related invasive infection unless data become available. Uptake into guideline recommendations will require larger, rigorous, and strain-specific trials.
Funding and trial registration
The Cochrane review update reported no funding source. The original review protocol (Johnston et al.) guided methods; the authors updated search strategy and selective reporting risk assessments for this iteration.
Selected references
1) Esmaeilinezhad Z, Ghosh NR, Walsh CM, Steen JP, Burgman AM, Mertz D, Johnston BC. Probiotics for the prevention of Clostridioides difficile-associated diarrhea in adults and children. Cochrane Database Syst Rev. 2025 Sep 11;9(9):CD006095. doi: 10.1002/14651858.CD006095.pub5.
2) McDonald LC, Gerding DN, Johnson S, et al. Clinical Practice Guidelines for Clostridium difficile Infection in Adults and Children: 2017 Update by the IDSA and SHEA. Clin Infect Dis. 2018;66(7):987–994. (Guideline document on C. difficile management and infection prevention.)
3) Lessa FC, Mu Y, Bamberg WM, et al. Burden of Clostridium difficile Infection in the United States. N Engl J Med. 2015;372:825–834. (Epidemiology and burden of disease.)
Practical tips for clinicians
– Review local baseline CDAD incidence before implementing routine probiotic prophylaxis; the absolute benefit scales with baseline risk.
– Use probiotic products with evidence from randomized trials when available and follow manufacturer storage/administration instructions to preserve viability.
– Avoid probiotic prophylaxis in patients who are severely immunocompromised or have central venous catheters unless robust safety data support use.
– Continue core infection-prevention strategies (antibiotic stewardship, hand hygiene, environmental cleaning) which remain the primary means to reduce CDAD.