In summary, for patients who took probiotics with antibiotics:

Benefits in NNT

    2.5% lower risk of C. difficile–associated diarrhea

    8.3% lower risk in high risk patients

    1 in 42 for preventing C. difficile–associated diarrhea (1 in 12 in high risk patients)

Harms in NNT

    No significant harm
    No significant harm

View As:

Source: Goldenberg JZ, Yap C, Lytvyn L, et al. Probiotics for the prevention of Clostridium difficile-associated diarrhea in adults and children. Cochrane Database Syst Rev 2017;12: CD006095

Study Population: 9,955 adults (>18 years) and children (0 to 18 years of age) from 39 trials receiving antibiotic therapy for any reason

Efficacy Endpoints: Clostridium difficile–associated diarrhea, probiotic adverse events, antibiotic-associated diarrhea, length of hospital stay, hospital survival, and 30-day survival

Harm Endpoints: Abdominal cramping, flatulence, soft stools, nausea, fever, and taste disturbance

Narrative: Clostridium difficile colitis is an opportunistic infection that occurs in individuals whose normal gut microbiota has been disrupted. Antibiotics can disturb the normal intestinal microflora, thereby reducing pathogen resistance to the Gram-positive, anaerobic, spore-forming bacillus. When a person takes antibiotics, good germs that protect against infection are destroyed for several months.1 During this time, patients can get sick from C. difficile, a Gram-positive anaerobic spore-forming bacillus picked up from contaminated surfaces or spread from a health care provider’s hands.1 C. difficile infection incidence in the United States has increased dramatically since 2000. In the United States, hospitalizations for C. difficile infection among nonpregnant adults doubled from 2000 through 2010 and were projected to continue to increase in 2011 and 2012.2 On the basis of data from U.S. death certificates, C. difficile infection is the leading cause of gastroenteritis-associated death and was estimated to cause 14,000 deaths in 2007.1, 2 Although almost half of infections occur in people younger than 65, more than 90% of deaths occur in those 65 and older.1 C. difficile has become the most common cause of health care–associated infections in U.S. hospitals, and the excess health care costs related to C. difficile infection are estimated to be as much as $4.8 billion for acute care facilities alone.2 All classes of antibiotics have been associated with C. difficile infections, and elevated risk may continue for up to 3 months postcessation. The spectrum of C. difficile–related disease varies from asymptomatic intestinal colonization, diarrhea, colitis, and pseudomembranous colitis with toxic megacolon.3

Probiotics are live organisms (bacteria or yeast) thought to confer a health benefit by improving host microbial balance, counteracting potential disturbances in intestinal flora associated with antibiotic use, and reducing the risk of colonization by pathogenic bacteria.4 They may prevent C. difficile-associated diarrhea (CDAD) through immune system stimulation, impact on the intestinal microbiome, direct inhibition of C. difficile growth, and toxin neutralization.4

The Cochrane meta-analysis cited here assessed the effectiveness of probiotics for decreasing the risk of CDAD among participants taking antibiotics.5 They included 39 randomized clinical trials (RCTs; 9,955 participants).5 A complete case analysis (i.e., participants who completed the study) among trials investigating CDAD (31 trials, 8,672 participants) suggests that probiotics reduce the incidence of CDAD when taken with antibiotics (RR = 0.4; 95% confidence interval [CI] = 0.3–0.52; absolute risk difference [ARD] = 2.4%; NNT = 42; GRADE quality of evidence, moderate). In other words, among patients receiving antibiotics, one would need to treat 42 patients with probiotics to prevent one case of CDAD.

The authors of the Cochrane meta-analysis performed a post hoc subgroup analysis on baseline risk of CDAD (low 0% to 2%; moderate 3% to 5%; high > 5%), determined by the event risk in placebo or no treatment group.5 Among studies with a baseline risk > 5%, the benefit of probiotics was even more pronounced (13 trials, 2,454 participants; RR = 0.30; 95% CI = 0.21–0.42; ARD = 8.1%; NNT = 12; GRADE quality of evidence, moderate). This means that one would need to treat 12 patients at high risk for CDAD with probiotics to prevent one case of CDAD resulting from antibiotic use.

In the Cochrane meta-analysis, 15 studies (n = 1,214) reported on the secondary endpoint of detection of C. difficile independent of presence of CDAD.5 Of these, 13 were placebo controlled and two trials used a no treatment control arm. The overall pooled results using a complete case approach did not show a statistically significant reduction in detection of C. difficile in the stool. C. difficile was detected in 15.5% (98/633) of the probiotics group compared to 17.0% (99/581) of the placebo or no treatment control group (RR = 0.86; 95% CI = 0.67–1.10; random effects).5 Eleven of the 15 studies were rated as having a high or unclear risk of bias. No statistically significant heterogeneity was detected for this comparison (p = 0.88; I2 = 0%).5

Caveats: This systematic review and meta-analysis assesses the safety and efficacy of probiotics for decreasing the risk of CDAD in adults and children taking antibiotics.5 Unfortunately, the analysis does not control for type or strain of the probiotics or for antibiotic class or administration route. In addition to the 39 trials (9,955 participants) identified by the Cochrane group, we identified three small relevant randomized clinical trials that were published after completion of the Cochrane review.6, 7, 8 These trials each studied a specific type of probiotic, exclusively in admitted patients, but did not show any significant difference for risk of CDAD between groups. They did not control for baseline risk and likely suffered from Type II error.6, 7, 8

Unfortunately, the meta-analysis does not provide a clear answer regarding the safety of probiotics. Adverse events (AEs) were a composite endpoint including pooled data on abdominal cramping, flatulence, soft stools, nausea, fever, and taste disturbance. The data for AEs are reported to be of low quality by the Cochrane review. Although significantly fewer AEs were reported in the probiotic compared to the control group (14.3% vs. 17.0%; RR = 0.83; 95% CI = 0.71–0.97), moderate heterogeneity was detected (p = 0.005; I2 = 49%).5 Additionally, 18 (56%) studies were rated as having a high or unclear risk of bias.

Conclusions are limited by the degree of missing data in the included trials. Moreover, the meta-analysis controlled for neither probiotic nor antibiotic types. In this manner one assumes that all probiotic treatments and antibiotic exposures are equivalent in their effect and that any variation in effect is due to chance.5 However, it is unknown if all probiotics are equally efficacious, if specific dosing thresholds need be met, or if effective probiotic dose varies based on antibiotic class being administered. Indeed, a meta-analysis by Allen et al.9 found that a formulation of Lactobacillus acidophilus + Lactobacillus casei + Lactobacillus rhamnosus showed positive results, whereas Saccharomyces boulardii alone and a combination of Lactobacillus acidophilus + Bifidobacterium bifidum + Bifidobacterium lactis was shown not to be effective.9

Finally, the Cochrane review included a mix of inpatients and outpatient subjects.5 However, the subgroup analysis did not show any significant impact for hospitalization. Regardless of this finding, it is plausible that patients admitted to the hospitals or intensive care units might respond differently to probiotics.

The green color recommendation is based on evidence of moderate quality data supporting benefits from probiotics in preventing CDAD in patients on antibiotics, in the absence of evidence of serious harms associated with their use.

Conclusion: Based on this systematic review and meta-analysis of 31 randomized controlled trials including 8,672 patients, moderate certainty evidence suggests that probiotics are effective for preventing CDAD. Although they did not describe how risk was determined, post hoc subgroup analyses indicated that probiotics are effective among trials with a CDAD baseline risk > 5% (moderate certainty evidence), but not among trials with a baseline risk ≤ 5% (low to moderate certainty evidence). No significant harm was reported (low-quality evidence). The effect of probiotic type, dose, antibiotic, and antibiotic route remains unclear.

Author: Ahmed Hamed, MD; Andrew C. Miller, MD

Published/Updated: November 5, 2018

  1. The Title Bar

    The title bar is color-coded with our overall recommendation.

    • Green: Benefits outweigh risks.
    • Yellow: Unclear risk/benefit profile.
    • Red: Benefits do not outweigh risks.
    • Black: Obvious harms, no clear benefits.
  2. Tip content...