Do probiotics or synbiotics taken during or after antibiotics help?

What it is

Probiotics are live microbial preparations given in defined doses; the trials in this area mostly used Lacticaseibacillus (Lactobacillus) rhamnosus GG, Saccharomyces boulardii, Bifidobacterium species, Bacillus species, Clostridium butyricum and mixtures of these. Synbiotics combine such organisms with a fermentable substrate. The outcome measured in almost all trials is antibiotic-associated diarrhoea, not microbiome restoration.

What the research says

Pooled randomised evidence shows fewer episodes of antibiotic-associated diarrhoea when a probiotic is taken alongside antibiotics, with the effect largest at higher doses and clearest for L. rhamnosus GG and S. boulardii. Certainty is low to moderate, many trials have company involvement, and one invasive human study found that a multi-strain probiotic delayed recovery of a person's own gut community compared with no supplement.

Evidence grade: Limited evidence.

What goes wrong

A multi-strain probiotic taken after antibiotics delayed and left incomplete the return of participants' own gut communities. (Source 1)

• Randomized trial, Low certainty.
• Size: human participants sampled by endoscopy plus mouse experiments (exact human n not stated in the abstract)
• Who: adults given antibiotics, and mice.
• How long: days to weeks after antibiotics.
• Result: Markedly delayed and persistently incomplete stool and mucosal microbiome reconstitution versus spontaneous recovery.
• Funding: not stated.

Collectively, potential post-antibiotic probiotic benefits may be offset by a compromised gut mucosal recovery

Bloodstream infection with Saccharomyces was recorded in 0.11% of hospitalised patients given S. boulardii, and intensive care admission was associated with more than a sixfold increase in that risk. (Source 2)

• Cohort study, Low certainty.
• Size: 46,729 hospitalised adults analysed, 16,404 given the probiotic.
• Who: adults admitted to inpatient medical units for at least 48 hours, 2016-2019.
• How long: 4 years.
• Result: Fungemia in 18 of 16,404 probiotic recipients (0.11%); 1.70 cases per 10,000 patient-days; 0.26 cases per 1,000 central-line days; ICU admission OR 6.55, 95% CI 2.28 to 18.87, incidence rate 0.47 cases per 1,000 patient-days.
• Funding: not stated.

S cerevisiae fungemia was identified in 18 probiotic recipients (0.11%). The observed incidence of fungemia attributable to S boulardii administration is 1.70 cases per 10,000 patient-days. Central-line days numbered 52,949 yielding an incidence of 0.26 cases of S cerevisiae per 1,000 central-line days. Intensive care unit admission was significantly associated with an increase in the risk of S cerevisiae (OR 6.55, CI 2.28-18.87), incidence rate of 0.47 cases per 1,000 patient-days.

The paediatric Cochrane review recorded no serious adverse events in its trials but pointed to serious events reported outside them in vulnerable children. (Source 3)

• Systematic review, Low certainty.
• Size: 4,415 children across 24 trials reporting adverse events.
• Who: children receiving antibiotics, including intensive care and neonatal units.
• How long: 5 days to 4 weeks.
• Result: Adverse events 4% (86/2229) with probiotics versus 6% (121/2186) with control; RD 0.00, 95% CI -0.01 to 0.01.
• Funding: not stated.

observational studies not included in this review have reported serious adverse events in severely debilitated or immuno-compromised children with underlying risk factors including central venous catheter use and disorders associated with bacterial/fungal translocation.

What the evidence supports

A 2025 Cochrane review found probiotics reduced antibiotic-associated diarrhoea, with low-certainty evidence. (Source 4)

• Systematic review, Low certainty.
• Size: 13,419 participants across 40 trials.
• Who: adults and children receiving antibiotics for any reason.
• How long: trial durations varied; follow-up to hospital discharge or weeks after treatment.
• Result: AAD in 23% (1592/6901) of probiotic participants versus 27.4% (1787/6518) of controls; RR 0.67, 95% CI 0.57 to 0.78; P < 0.001.
• Funding: no funding for the review update; 28 of the included studies had probiotic-company affiliation or support.

Limit of this finding: Hold one number in this quotation at arm's length. The review calls the result 'a large ARR of 9%', but its own incidence figures, 23% against 27.4%, differ by 4.4 percentage points, not 9. The relative risk of 0.67 (95% CI 0.57 to 0.78) is the figure the analysis actually supports, and the review rates it low certainty. You should conclude that probiotics probably reduced antibiotic-associated diarrhoea somewhat; you should not take 9% as the size of that reduction.

Pooled results from 40 studies (13,419 participants) suggest that probiotics may result in a large ARR of 9% (incidence of 23% [1592/6901] in the probiotic group versus 27.4% [1787/6518] in the control group), and an RRR of 33% (RR 0.67, 95% CI 0.57 to 0.78; P < 0.001; low-certainty evidence) in incidence of AAD.

A 2019 Cochrane review in children found probiotics roughly halved antibiotic-associated diarrhoea, with the benefit concentrated in higher-dose trials. (Source 5)

• Systematic review, Moderate certainty.
• Size: 6,352 children across 33 trials.
• Who: children aged 0 to 18 receiving antibiotics.
• How long: follow-up 5 days to 12 weeks.
• Result: AAD 8% (259/3232) with probiotics versus 19% (598/3120) with control; RR 0.45, 95% CI 0.36 to 0.56; NNTB 9, 95% CI 7 to 13; high-dose trials (at least 5 billion CFU/day) RR 0.37, 95% CI 0.30 to 0.46.
• Funding: not stated in the abstract; 20 of 33 trials at high risk of bias.

Limit of this finding: One caution about the source: in the same Cochrane abstract the p-values printed for the two dose subgroups contradict their own confidence intervals ('P = 0.06' beside a clearly significant high-dose result, 'P = 0.02' beside a low-dose interval that crosses 1.00). The main result quoted here is unaffected, but the subgroup p-values from this abstract should not be repeated as published.

After 5 days to 12 weeks of follow-up, the incidence of AAD in the probiotic group was 8% (259/3232) compared to 19% (598/3120) in the control group (RR 0.45, 95% CI 0.36 to 0.56; I² = 57%, 6352 participants; NNTB 9, 95% CI 7 to 13; moderate certainty evidence).

In outpatients, pooled trials of L. rhamnosus GG and S. boulardii showed lower rates of antibiotic-associated diarrhoea. (Source 6)

• Meta-analysis, Moderate certainty.
• Size: 3,631 participants across 17 randomised controlled trials.
• Who: outpatients receiving antibiotics.
• How long: varied by trial.
• Result: AAD 8.0% with probiotics versus 17.7% with control; RR 0.49, 95% CI 0.36 to 0.66; I2 = 58%.
• Funding: not stated.

Limit of this finding: A number in this abstract is mis-set. The adverse-event analysis is reported as covering '2.363 participants', which is a European decimal comma turned into a point: it means 2,363 people, not 2.363. The review's 17 trials include 3,631 participants in total. The figure is a typographical fault in the source, not a finding about a fraction of a person.

The pooled results found that AAD was present in 8.0% of the probiotic group compared to 17.7% in the control group (RR 0.49, 95% CI 0.36 to 0.66; I2 = 58%), and the species-specific results were similar regarding the probiotic strains L. rhamnosus GG and S. boulardii.

A meta-analysis in adults reported a 38% relative reduction in antibiotic-associated diarrhoea. (Source 7)

• Meta-analysis, Low certainty.
• Size: 9,312 participants across 36 randomised placebo-controlled trials.
• Who: adults receiving antibiotics.
• How long: varied by trial.
• Result: Pooled RR 0.62, 95% CI 0.51 to 0.74; adverse events RR 1.00, 95% CI 0.87 to 1.14.
• Funding: not stated.

Probiotics reduced the incidence of AAD by 38% (pooled relative risk, 0.62; 95% confidence interval, 0.51-0.74).

What the evidence does not support

Probiotics did not significantly reduce carriage of C. difficile in stool in people without symptoms. (Source 4)

• Systematic review, Low certainty.
• Size: 1,302 participants across 16 trials.
• Who: adults and children receiving antibiotics.
• How long: trial durations varied.
• Result: Colonization 14.9% (101/677) with probiotics versus 16.2% (101/625) with control; RR 0.87, 95% CI 0.68 to 1.11; P = 0.27.
• Funding: no funding for the review update.

Limit of this finding: The Cochrane abstract's own arithmetic does not hold together for this result. It calls it an absolute risk reduction of 2.1%, but the two figures it gives, 14.9% and 16.2%, differ by 1.3 percentage points, and it prints the same numerator, 101, over two different denominators (101/677 and 101/625). Read the risk ratio instead: 0.87, with a confidence interval from 0.68 to 1.11 that crosses 1, meaning no difference was shown. You should conclude that probiotics were not shown to reduce carriage of C. difficile; you should not treat the 2.1% figure or the paired counts as reliable numbers.

RR 0.87, 95% CI 0.68 to 1.11; P = 0.27; low-certainty evidence

In the paediatric Cochrane review, trials using less than 5 billion CFU per day did not show a statistically significant reduction in antibiotic-associated diarrhoea. (Source 5)

• Systematic review, Low certainty.
• Size: 2,214 children in low-dose trials.
• Who: children receiving antibiotics.
• How long: follow-up 5 days to 12 weeks.
• Result: Low-dose trials: 8% (97/1155) versus 13% (133/1059); RR 0.68, 95% CI 0.46 to 1.01.
• Funding: not stated.

Limit of this finding: The p-value in this quotation is wrong in the source. The Cochrane abstract prints 'P = 0.02' for this low-dose result even though the confidence interval it gives alongside, 0.46 to 1.01, includes 1.00 and therefore means the result is not statistically significant. The same abstract prints 'P = 0.06' beside the high-dose result, which is plainly significant; the two appear to have been swapped or mislabelled. You should read these subgroups from their confidence intervals; you should not read the printed p-values as telling you which doses worked.

For the low dose studies the incidence of AAD in the probiotic group was 8% (97/1155) compared to 13% (133/1059) in the control group (2214 participants; RR 0.68; 95% CI 0.46 to 1.01; P = 0.02).

Probiotics made no meaningful difference to length of hospital stay. (Source 4)

• Systematic review, Moderate certainty.
• Size: 6,553 participants across 7 trials.
• Who: hospitalised patients receiving antibiotics.
• How long: length of admission.
• Result: MD -0.07 days, 95% CI -0.35 to 0.21; P = 0.63.
• Funding: no funding for the review update.

Limit of this finding: This particular result is internally consistent, but two other results in the same Cochrane abstract are not: the passages on C. difficile carriage and on antibiotic-associated diarrhoea each state an absolute risk reduction that does not match their own percentages. Figures taken from elsewhere in this abstract should be judged by the risk ratios and confidence intervals rather than by the absolute percentages printed beside them.

There is likely little to no difference in length of hospital stay between treatment and control groups (MD -0.07, 95% CI -0.35 to 0.21; P = 0.63; moderate-certainty evidence).

An umbrella review of 20 systematic reviews in children rated most of that review literature low or critically low quality. (Source 8)

• Review of reviews, Low certainty.
• Size: 20 systematic reviews; 47 GRADE-rated outcomes.
• Who: children with antibiotic-associated diarrhoea.
• How long: reviews published to October 2022.
• Result: 3 reviews moderate quality, 10 low, 7 critically low on AMSTAR 2; 24 of 47 outcomes low-quality and 4 critically low on GRADE.
• Funding: not stated.

the results of AMSTAR 2 showed that 3 SRs belonged to moderate quality level, 10 SRs belonged to low-quality level and 7 SRs being extremely low-quality level

Where the research disagrees

Whether probiotics taken with antibiotics help or hinder the gut microbiome

• Cochrane review authors (Goldenberg and colleagues, 2025 update), systematic review of 47 randomised trials with GRADE ratings (low certainty for CDAD): The currently available evidence suggests that probiotics may be effective for preventing CDAD, suggesting that for every 65 people taking probiotics, one case of CDAD may be prevented (Source 9)
• Suez and colleagues (Cell, 2018), invasive human sampling with mouse experiments, microbiome endpoints rather than clinical diarrhoea: Collectively, potential post-antibiotic probiotic benefits may be offset by a compromised gut mucosal recovery (Source 1)

A common belief, and what the research shows

The belief: Probiotics restore the gut microbiome after antibiotics.

What the research shows: The trials that show benefit measured diarrhoea, not restoration, and the one study that looked directly at reconstitution found the opposite: "probiotics induced a markedly delayed and persistently incomplete indigenous stool/mucosal microbiome reconstitution and host transcriptome recovery toward homeostatic configuration, while aFMT induced a rapid and near-complete recovery within days of administration."

Questions and answers

Probiotics are live microbial preparations given in defined amounts, taken alongside or after antibiotics in the hope of preventing diarrhoea and dysbiosis. Synbiotics pair such organisms with a fermentable substrate. The trials pooled by Cochrane used any strain or dose. (Source 10)

In pooled randomised trials, taking a probiotic with antibiotics lowered the rate of antibiotic-associated diarrhoea from about 27% to about 23% of participants, a relative reduction of a third, rated low certainty. In children the pooled reduction was larger. (Source 4)

For diarrhoea prevention the balance in the trials is favourable and adverse events were no more common than in controls. For microbiome restoration the picture is different: an invasive human study found that a multi-strain probiotic delayed the return of participants' own gut community after antibiotics. (Source 1)

Reported as studied, not as advice: the paediatric Cochrane subgroup analysis compared trials giving at least 5 billion CFU per day with trials giving less, and the benefit was concentrated in the higher-dose trials. Outpatient pooled results were similar for L. rhamnosus GG and S. boulardii. (Source 5)

Not applicable in the usual sense: these organisms are taken deliberately and are largely transient. The relevant finding is that antibiotic disruption made the human mucosa more, not less, colonisable by the supplement strains. (Source 1)

This question is about being low in a nutrient or resident organism; probiotic strains are supplements rather than residents. What the reviews describe is the antibiotic disturbance the supplements are meant to address. (Source 10)

Fermented foods are the traditional source of these organisms, described in a narrative review rather than tested head-to-head against capsules in the antibiotic trials we retrieved. Doses and strains in fermented foods are not standardised. (Source 11)

Without a probiotic, about 27% of trial participants taking antibiotics developed diarrhoea and about 3.2% developed C. difficile-associated diarrhoea in the control arms of the Cochrane trials. (Source 4)

There is no validated test that tells an individual whether a probiotic has worked. Trials measure clinical diarrhoea, and stool detection of C. difficile was measured separately because carriage without symptoms is not the same as disease. (Source 4)

We searched: Cochrane 2025 CDAD review, Cochrane 2019 paediatric AAD review, 2023 umbrella review of paediatric reviews; none reported a validated individual test of probiotic effect.

References

1. Cell. Post-Antibiotic Gut Mucosal Microbiome Reconstitution Is Impaired by Probiotics and Improved by Autologous FMT.. 2018. PMID 30193113, DOI 10.1016/j.cell.2018.08.047. Read the source
2. Mycoses. Incidence of Saccharomyces cerevisiae fungemia in hospitalised patients administered Saccharomyces boulardii probiotic. 2021. DOI 10.1111/myc.13375. Read the source
3. Cochrane Database of Systematic Reviews. Probiotics for the prevention of pediatric antibiotic-associated diarrhea.. 2019. PMID 31039287, DOI 10.1002/14651858.cd004827.pub5. Read the source
4. Cochrane Database of Systematic Reviews. Probiotics for the prevention of Clostridioides difficile-associated diarrhea in adults and children.. 2025. PMID 40931979, DOI 10.1002/14651858.cd006095.pub5. Read the source
5. Cochrane Database of Systematic Reviews. Probiotics for the prevention of pediatric antibiotic-associated diarrhea.. 2019. PMID 31039287, DOI 10.1002/14651858.cd004827.pub5. Read the source
6. Antibiotics. Probiotics for the Prevention of Antibiotic-Associated Diarrhea in Outpatients—A Systematic Review and Meta-Analysis. 2017. DOI 10.3390/antibiotics6040021. Read the source
7. Journal of Clinical Gastroenterology. Probiotics for the Prevention of Antibiotic-associated Diarrhea in Adults. 2021. DOI 10.1097/mcg.0000000000001464. Read the source
8. Frontiers in Pharmacology. Overview of systematic reviews of probiotics in the prevention and treatment of antibiotic-associated diarrhea in children. 2023. DOI 10.3389/fphar.2023.1153070. Read the source
9. Cochrane Database of Systematic Reviews. Probiotics for the prevention of Clostridioides difficile-associated diarrhea in adults and children.. 2025. PMID 40931979, DOI 10.1002/14651858.cd006095.pub5. Read the source
10. Cochrane Database of Systematic Reviews. Probiotics for the prevention of Clostridioides difficile-associated diarrhea in adults and children.. 2025. PMID 40931979, DOI 10.1002/14651858.cd006095.pub5. Read the source
11. Healthcare (narrative review). Probiotics for the Prevention of Antibiotic-Associated Diarrhea. 2022. DOI 10.3390/healthcare10081450. Read the source