Should Probiotics be Routine Therapy for the Prevention of Antibiotic-associated Diarrhea?

Accumulating evidence suggests that disturbances in the intestinal microbiota underlie many diseases and that therapeutic manipulation of microbial balance has the potential to ameliorate disease symptoms.1 Microbial balance in the gut can be adjusted through the use of antibiotics, probiotics, and prebiotics. Antibiotics offer one common avenue for improving health through suppression and elimination of microbial pathogens. However, antibiotic-mediated disruption of intestinal microbial balance can result in a variety of undesirable side effects, most notably antibiotic-associated diarrhea (AAD). AAD is a common problem occurring in 5% to 25% of patients receiving antibiotics, with rates varying depending on the population studied and the antibiotic used.2 Infection with Clostridium difficile, the most commonly diagnosed and potentially severe form of AAD, accounts for approximately 25% of AAD. Probiotics given in conjunction with antibiotics have been extensively studied for the prevention of AAD in both adults and children.3 Five meta-analyses of trials conducted using a variety of different probiotics for the prevention of AAD have been performed.4–8 Although the trials analyzed in these meta-analyses evaluated different probiotic organisms, doses, durations of therapy, endpoints, and study populations, the individual meta-analyses each showed an overall reduction in the risk of AAD during probiotic administration with antibiotics. Importantly, no significant treatment-related side effects were found. In particular, treatment with Lactobacillus rhamnosus GG (LGG) and Saccharomyces boulardii seemed to have the most significant efficacy,8 although these were also the most commonly investigated probiotics. A recent rigorous review of the literature on the prevention of pediatric AAD from the Cochrane Database of Systematic Reviews concluded that the existing data was promising, particularly for LGG and S. boulardii, but not conclusive.9 This review reported a number needed to treat with probiotics to prevent one case of diarrhea in children treated with antibiotics of 10. Although probiotic therapy decreased the duration of diarrhea by 0.75 days, there was no significant impact on diarrhea frequency.9 Studies that used doses of probiotics of at least 5 billion CFUs/day were more likely to show a beneficial effect, as were studies of LGG or S. boulardii. Finally, funnel plot analysis suggested a possible publication bias favoring the reporting of positive studies with few negative or inconclusive studies in the literature.9 In this issue of the Journal, Lönnermark and colleagues10 present the results of a randomized, placebo controlled trial of another probiotic, Lactobacillus plantarum 299v, for the prevention of AAD in adults. L. plantarum 299v is known to survive passage through the gastrointestinal tract, can be isolated from rectal biopsies, and has been shown to reduce symptoms in patients with irritable bowel syndrome, qualities making it a reasonable probiotic for study in patients at risk for AAD. The authors enrolled both inpatients and outpatients above 16 years of age within 48 hours of starting antibiotic therapy. Patients randomly received a drink containing 1010 CFU of L. plantarum 299v or an identical-tasting placebo drink. Participants ingested the test drink during the time they were being treated with antibiotics, for 1 week after completion of antibiotic therapy, and were then followed an additional week. The primary study outcome was the proportion of study subjects in each group developing diarrhea, defined as at least 3 loose or watery stools per day for at least 2 consecutive days, and positive for C. difficile toxin after antibiotic treatment. Two hundred thirty-nine subjects enrolled in the trial, however, only 163 (80 L. plantarum and 83 placebo) completed the protocol, although the reasons for withdrawal did not differ between the 2 groups. Unfortunately, AAD, the primary trial outcome, was uncommon in both the treatment (6%) and placebo (7.5%) group during the entire study period. Most trials of probiotics for AAD report diarrhea in around 15% to 20% of placebo-treated patients, with around 25% to 50% reduction in diarrhea with probiotic therapy compared with placebo.11 Given the low occurrence of diarrhea in this study, no benefit of L. plantarum 299v for the prevention of AAD could be shown. However, treatment with L. plantarum 299v was associated with a decreased risk of developing at least one episode of loose or watery stools compared with placebo treatment with an odds ratio of 0.69 (95% confidence interval 0.52-0.92, p=0.012). The benefit was most noticeable when subjects were receiving both antibiotics and the study drink or had completed antibiotic therapy, but were still receiving the study drink. Nausea was also reduced in the treatment group as well. Although the authors were not able to show an overall benefit of probiotic treatment on AAD, they recommended further trials of L. plantarum 299v given the observed reduction in the incidence of antibiotic-associated loose stools.10 Any further studies with L. plantarum 299v, however, will need to be better designed and better executed to give the best chance of a meaningful outcome. Like many trials of probiotics in AAD,9 no power calculations were made for this trial before study initiation, so from the start we do not know whether a type II error is likely. The very high dropout rate of above 31% is worrisome, suggesting that the investigators were either not able to adequately motivate patients to remain in the trial or effectively track patients once they were enrolled. The trial looked at a fairly heterogeneous group of patients with a wide age range, a wide variety of infectious conditions, different antibiotic regimens, and both inpatient and outpatient status. It might be too much to ask for clinical significance in a relatively small trial with so many variables at work. Finally, as with most probiotics, no data is available on the optimal dose of L. plantarum 299v as dose ranging trials have not been performed. Overall, however, the preponderance of the evidence strongly supports the use of probiotics for the prevention of AAD.12 Although the beneficial effects of probiotics are neither profound nor perfect, these therapies are safe for most of the patients, an argument supporting their routine use.11 Importantly, probiotics for the prevention of AAD appear to be very cost-effective. A recent trial of a probiotic yoghurt drink reported a cost to prevent one case of AAD of $100 and a cost of $120 to prevent one case of C. difficile diarrhea.13 For hospitalized patients, estimates suggest a per patient cost of $3669 to treat C. difficile in the United States.14 Even with modest efficacy, the potential cost savings associated with the routine use of probiotic therapy for the prevention of AAD are profound. Still, despite the numerous clinical trials, many questions surrounding the use of probiotics for the prevention of AAD remain and more and better data is needed before the probiotic use is routine. Future studies will need to focus on the most pressing issues requiring investigation. Not only do we need to better understand questions of probiotic mechanisms of action, choice of probiotic or combination of probiotic, dose, and duration of therapy, but we also need to focus our efforts on the most relevant clinical issues. Although AAD can be an unpleasant and annoying side effect of antibiotic therapy, in adults it is usually unlikely to be clinically significant. However, preventing C. difficile infection, especially in nursing homes and among hospitalized patients has profound implications for our patients' health and our nation's healthcare. To provide definitive answers these important questions, we will need to move away from smaller, poorly designed or poorly executed single institution studies, and instead perform adequately powered, well-designed, well-funded, multi-institutional trials. Although Lönnermark and colleagues fell short in their efforts to answer some of these questions, the future for probiotic therapy in the prevention of AAD remains an exciting area for investigation where much work remains to be carried out.