Probiotics for the prevention of Clostridioides difficile-associated diarrhea in adults and children.

RATIONALE

Antibiotics can disturb gastrointestinal microbiota, which may lead to reduced resistance to pathogens such as Clostridioides difficile. Probiotics are live microbial preparations that, when administered in adequate amounts, may confer a health benefit to the host, and are a potential C difficile infection prevention strategy.

OBJECTIVES

To assess the benefits and harms of probiotics for preventing C difficile-associated diarrhea (CDAD) in adults and children receiving an antibiotic for any reason.

SEARCH METHODS

We searched the Cochrane Central Register of Controlled Trials (CENTRAL), Ovid MEDLINE, and Ovid Embase from 18 March 2017 (search date of the previous version of the review) to 3 March 2025.

ELIGIBILITY CRITERIA

We included randomized controlled trials with a placebo, alternative prophylaxis, or no-treatment control investigating probiotics (any strain or dose) for the prevention of CDAD in adults and children receiving an antibiotic for any reason.

OUTCOMES

The critical outcome was incidence of CDAD. Important outcomes included detection of C difficile in stool (i.e. colonization in absence of symptoms), adverse events, antibiotic-associated diarrhea (AAD), and length of hospital stay.

RISK OF BIAS

Two review authors independently assessed risk of bias using Cochrane's RoB 1 tool.

SYNTHESIS METHODS

We pooled dichotomous outcomes (e.g. incidence of CDAD) using a random-effects complete-case model to calculate the risk ratio (RR) and absolute risk reduction (ARR) and corresponding 95% confidence interval (95% CI). We pooled continuous outcomes (e.g. length of hospital stay) using a random-effects model to calculate the mean difference (MD) and corresponding 95% CI. We conducted sensitivity analyses to explore the impact of missing outcome data on benefits and harms outcomes. To investigate heterogeneity, we conducted a priori subgroup analyses on risk of bias, probiotic dose, probiotic species, adults versus children, inpatients versus outpatients, and the event rate in the control group for developing CDAD (low 0% to 2%; moderate 3% to 5%; high > 5%). The certainty of the evidence for each outcome was independently assessed using the GRADE approach.

INCLUDED STUDIES

Eight new studies (n = 4595) met our inclusion criteria, for a total of 47 included studies (n = 15,260). Although we noted a high risk of bias in individual studies for sequence generation or allocation concealment, this did not impact our findings. Around 30% of studies lacked a published protocol or clinical trial registration, and in some cases there were inconsistencies between the reported methods and published results. We noted authors' affiliation with a probiotic company and financial support received from probiotic companies in 28 studies.

SYNTHESIS OF RESULTS

Evidence from 38 trials (13,179 participants) suggests that probiotics may result in a small absolute risk reduction (ARR) of 1.6% (incidence of 1.6% [110/6787] in the probiotic group versus 3.2% [203/6392] in the control group), and a relative risk reduction (RRR) of 50% (RR 0.50, 95% CI 0.38 to 0.64; P < 0.001; low-certainty evidence) in incidence of CDAD. Twenty-seven of 38 trials had missing CDAD data, ranging from 2% to 45%. Our complete-case CDAD results proved robust to sensitivity analyses of plausible and worst-plausible assumptions regarding missing outcome data. Pooled results from 16 trials (1302 participants) suggest that probiotics may result in a small absolute reduction in C difficile in stool (i.e. colonization in absence of symptoms) (ARR 2.1%; incidence of 14.9% [101/677] in the probiotic group versus 16.2% [101/625] in the control group; RR 0.87, 95% CI 0.68 to 1.11; P = 0.27; low-certainty evidence). Thirty-seven studies (11,911 participants) reported on adverse events. Our pooled analysis indicates that probiotics probably result in a small ARR of 1.7% (incidence of 10.9% [669/6148] in the probiotic group versus 12.2% [701/5763] in the control group), and an RRR of 14.0% (RR 0.86, 95% CI 0.72 to 1.01; P = 0.074; moderate-certainty evidence). The most common adverse events in both the treatment and control groups were abdominal cramping, nausea, fever, soft stools, flatulence, and taste disturbance. 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. Only seven trials (6553 participants) investigated length of hospital stay. 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). We downgraded the certainty of evidence for CDAD and AAD due to the low number of events or publication bias. We observed some variability in CDAD control group event rate (< 2% to > 5%) that may impact the applicability of evidence depending on clinical setting. We downgraded the certainty of evidence for adverse events due to variation between the results of the studies and for hospital stay due to selective reporting.

AUTHORS' CONCLUSIONS: 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 (number needed to treat for an additional beneficial outcome of 65, 95% CI 48 to 97). There are probably fewer adverse events in the probiotic group. The short-term use of probiotics may have a small absolute benefit, and is likely not associated with adverse effects when used in patient populations receiving systemic antibiotics who are not immunocompromised. Large trials comparing probiotics with placebo in people with a low baseline risk of CDAD are needed.

FUNDING

There was no source of funding for this update.

REGISTRATION

The original review was done in 2013 by Johnston and colleagues, for which the protocol was developed. We followed the methods specified in the 2017 review, while modifying the search strategy and risk of bias assessment for selective reporting for this update.