Key Points

Introduction

Diarrhea is one of the most frequent side effects of antibiotic use. The incidence of antibiotic-associated diarrhea (AAD) varies between 5% and 39% of patients, with a higher percentage seen in hospitalized patients.

AAD is defined as otherwise unexplained diarrhea that occurs in association with the administration of antibiotics. It is characterized by a change in the normal stool frequency with at least 3 loose or watery stools daily for 3 days. Early onset of diarrhea occurs within 2 to 7 days, being earlier with children than outpatient adults. Delayed onset of diarrhea may occur within 2 to 8 weeks after the antibiotic has been discontinued.

Special attention should be accorded to at-risk hospitalized patients who may develop Clostridium difficile infection (CDI) complicating AAD. Prompt stool analysis for toxins A and B is essential for these patients. At-risk patients for CDI include patients aged greater than 65 years, patients with multiple comorbidities, immunosuppression, exposure to radiation and chemotherapy, inflammatory bowel disease (IBD), hepatic cirrhosis, prolonged hospitalization, and treatment with proton pump inhibitors (PPIs) ( Box 1 ). Pharmacologically, the use of broad spectrum antibiotics and the duration of antibiotic therapy increase risk of AAD.

Introduction

Diarrhea is one of the most frequent side effects of antibiotic use. The incidence of antibiotic-associated diarrhea (AAD) varies between 5% and 39% of patients, with a higher percentage seen in hospitalized patients.

AAD is defined as otherwise unexplained diarrhea that occurs in association with the administration of antibiotics. It is characterized by a change in the normal stool frequency with at least 3 loose or watery stools daily for 3 days. Early onset of diarrhea occurs within 2 to 7 days, being earlier with children than outpatient adults. Delayed onset of diarrhea may occur within 2 to 8 weeks after the antibiotic has been discontinued.

Special attention should be accorded to at-risk hospitalized patients who may develop Clostridium difficile infection (CDI) complicating AAD. Prompt stool analysis for toxins A and B is essential for these patients. At-risk patients for CDI include patients aged greater than 65 years, patients with multiple comorbidities, immunosuppression, exposure to radiation and chemotherapy, inflammatory bowel disease (IBD), hepatic cirrhosis, prolonged hospitalization, and treatment with proton pump inhibitors (PPIs) ( Box 1 ). Pharmacologically, the use of broad spectrum antibiotics and the duration of antibiotic therapy increase risk of AAD.

Causes of AAD

Antibiotics cause diarrhea by several mechanisms. First, suppression of anaerobic bacteria results in reduced metabolism of carbohydrates inducing an osmotic diarrhea. Secondly, antibiotics disrupt the protective effect of commensal bacteria. The disruption of microbial diversity reduces colonic mucosal resistance to pathogenic opportunistic bacteria, particularly CDI. Following discontinuation of the antibiotic, restoration of the normal commensal bacteria may take several weeks, months, or longer to occur, thus placing the patient at longer term risk for disease-causing pathogenic agents. Finally, antibiotics with prokinetic activity, such as erythromycin and clavulanate, promote diarrhea.

Preventing AAD with probiotics

Multiple studies support the use of probiotics for preventing AAD. AAD presents a fertile area to study the efficacy of probiotics. The study design is simplified because diarrhea is a predictable side effect; the impact of the antibiotic usually occurs early in the course of administration; the duration of therapy is usually limited (eg, 10–14 days), and, in most instances, the side effects of added probiotics are minimal. In addition, given the spectrum of AAD, possibly eventuating to the more serious CDI, an effective probiotic represents a major step in reducing more serious illness. Most studies, both in children and adults, have been reported with positive results. Interpretation of studies is difficult partly because of the varied populations, age of the patients, comorbidities, drug interactions, duration of probiotic administration, numbers of patients involved, the nature and dose of the offending antibiotic agent, and the lack of randomized, placebo-controlled trials ( Box 2 ). Clinically acceptable probiotics must be species specific; of human origin; survive passage from the oral cavity through the gastric acid barrier, digestive enzymes, and bile acids; travel down the small bowel into the colon; nidate; and proliferate therein. Probiotics should be of adequate dose, preferably greater than 10 billion cfu/gm in adults, maintain their viability and concentration, and have a dependably measurable shelf life at the time of purchase and administration ( Box 3 ). When these qualities have been met, targeted illnesses require randomized, placebo-controlled, double-blinded trials on appropriate populations.

Mechanisms of action of probiotics

Probiotics offer protection from potential pathogens by enhancing mucosal barrier function by secreting mucins, providing colonization resistance, producing bacteriocins, increasing production of secretory immunoglobulin A (IgA), producing a balanced T-helper cell response, and increasing production of interleukin 10 (IL-10) and transforming growth factor beta, both of which play a role in the development of immunologic tolerance to antigens ( Box 4 ).

Single strain probiotics used for treating AAD

Multistrain probiotics for treating AAD

Prevention of pediatric AAD

Six placebo-controlled, RCTs comprising 766 children were included in a meta-analysis of probiotics preventing AAD that was published in 2006. Treatment with probiotics compared with placebo reduced the risk of AAD from 28.5% to 11.9%. Preplanned subgroup analysis showed that reduction of the risk of AAD was associated with the use of LGG, S. boulardii, or B. lactis, and Streptococcus thermophilus. Probiotics reduced the risk of AAD in children; of every 7 patients who would develop diarrhea while being treated with antibiotics, one fewer will develop AAD if also receiving probiotics.

No adverse effects were observed in any of the included trials. However, the investigators cautioned regarding the use of probiotics in immune-compromised patients. They recommended identification of populations at high risk of AAD who would benefit most from use of probiotic therapy, assessment of additional probiotic strains, designing an effective dosing regimen, and addressing the cost effectiveness of using probiotics to prevent AAD in children.

A Cochrane Database review of the data on probiotics in preventing pediatric AAD was published in 2011. Sixteen studies were reviewed and provided the best available evidence. The studies tested 3432 children (2 weeks to 17 years of age) who were receiving probiotics coadministered with antibiotics to prevent AAD. These short-term studies showed probiotics to be effective for preventing AAD. Probiotics were generally well tolerated (no significant side effects between probiotics and control groups). Both L. rhamnosus and S. boulardii at dosages of 5 to 40 billion CFU/d may prevent the onset of ADD, with no serious side effects in otherwise healthy children. This benefit for high-dose probiotics needs to be confirmed by a large randomized study. No conclusions regarding other probiotic agents could be drawn.

Preventing adult AAD

Compiling data referable to the efficacy of probiotics suppressing AAD in adults is complicated by numerous factors. Variations in the clinical setting (hospitalized or community), numbers and ages of patients, nature of the population, type of antibiotic, duration of therapy, use of different probiotics, failure to designate the strain, variable dosages, and duration of therapy are some of the major factors making statistical judgment difficult. Three meta-analyses have been published in an effort to obtain useful clinical data. In 2002, D’Souza and colleagues and Cremonini and colleagues combined 9 and 7 trials, respectively.

In the D’Souza study, 2 of the 9 studies investigated the effects of probiotics in children. Four trials used S. boulardii, 4 used lactobacilli, and one used a strain of enterococcus that produced lactic acid. Three trials used a combination of probiotic strains of bacteria. The probiotics were given in combination with antibiotics, whereas the control groups received placebo and antibiotics. The odds ratio (OR) in favor of active treatment over placebo was 0.39 (95% confidence interval [CI], 0.25–0.62, P <.001) for the yeast and 0.34 (0.19–0.61, P <.01) for lactobacilli. The combined OR was 0.37 (0.26–0.53; P <.001) in favor of active treatment over placebo.

The investigators concluded that biotherapeutic agents may be useful in preventing AAD. Cremonini and colleagues (vide supra) limited their meta-analysis to 7 trials, which used the 2 most widely used probiotics Lactobacillus spp. and S. boulardii. Two studies involved children; 3 assessed the decrease in the occurrence of AAD during the administration of S. boulardii, and 4 during the administration of Lactobacillus spp. The search was limited to randomized studies. The inclusion criteria included a placebo design, with diarrhea as a primary end-point and a minimum of 2 weeks of follow-up. A total number of 881 patients were studied. The combined relative risk (RR) was 0.3966 (95% CI, 0.27–0.57). The results suggested a strong benefit of probiotic administration on AAD.

Meta-analytic pooling

A large and comprehensive meta-analysis pooling trials of probiotics in the prevention of pediatric and adult AAD was accomplished by McFarland in 2006. Study selection included trials in which specific probiotics were given to either prevent or treat the diseases of interest.

Trials were required to be randomized, blinded, controlled in humans, and published in peer-reviewed journals. Thirty one of 180 screened studies, totaling 3164 subjects, met the inclusion and exclusion criteria. From 25 RCTs (2810 patients), probiotics significantly reduced the RR of AAD (RR = 0.43, 95% CI, 0.31–0.58, P <.001).

Six of the trials, which included 1119 patients, used S. boulardii as the probiotic. One of the trials involved the use of triple antibiotics with concomitant S. boulardii for the eradication of Helicobacter pylori (43 patients). The same probiotic was used in a 246 pediatric patient trial. Six trials used LGG as the single probiotic; 388 pediatric patients in 3 trials, 267 adult patients in one trial, and 262 patients involved in 3 triple antibiotic/LLG trials for the eradication of H. pylori.

Overall, of 16 RCTs of adult AAD, 7 (44%) showed efficacy, whereas 9 RCTs (67%) showed significant efficacy in children. Of note, optimal results occurred with dosages of probiotics greater than 10 ¹⁰ cfu/g. Sixteen of 31 (84%) trials reported on adverse effects. In 24 trials, there were no adverse episodes. Nine percent of patients on S. boulardii reported increased thirst and 14% reported increased constipation. Thirty seven percent of patients on LGG reported mild gaseousness and 25% noted bloating.

AAD and helicobacter pylori therapy

A recent review of S. boulardii included 3 trials conducted in patients with H. pylori infection. These patients were treated with triple therapy, which includes 2 antibiotics and a PPI for 2 weeks.

Duman and colleagues treated 204 patients with 1 g of S. boulardii (2 × 10 ¹⁰ /d for 2 weeks) and triple therapy compared with 185 patients who received only triple therapy. Of the 389 patients, 376 completed the treatment phase and the 4-week follow-up. Fewer patients given S. boulardii (6.9%) developed AAD compared with the control group (15.6%, P = .007).

Two other RCTs were conducted in adult patients receiving triple therapy for H. pylori infections and both showed a significant reduction in AAD for those treated with S. boulardii. In summary, the use of concomitant probiotic therapy in association with multiple antibiotics and PPIs is helpful in reducing the antibiotic-induced gastrointestinal side effects.

Meta-analytic confirmation of results supporting use of probiotics for AAD

A statistically sophisticated meta-analysis adding confirmation for the use of probiotics for treatment of AAD was published in April, 2012 by Videlock and Cremonini. They used a meta-analysis of randomized, double-blinded, placebo-controlled trials including patients treated with antibiotics and administered a probiotic for at least the duration of the antibiotic treatment.

Meta-analysis and meta-regression methods were used to synthesize data and to assess influence of mean age, duration of antibiotics, risk of bias, and incidence of diarrhea in the placebo group on outcomes. Subgroup analyses explored effects of different probiotic species, patient populations, and treatment indications.

Thirty four studies were included with 4138 patients. Pooled RR for AAD in the probiotic group versus placebo was 0.53 (95% CI, 0.44–0.63), corresponding to a number needed to treat (NNT) of 8 (95% CI, 7–11). Pooled RR for AAD during treatment of H. pylori was 0.37 (95% CI, 0.20–0.69), with an NNT of 5 (95% CI, 4–10). This updated meta-analysis confirmed the results enumerated, supporting the effects of probiotics in AAD. Studies on the effect of probiotics in preventing the occurrence or reoccurrence of Clostridium difficile colitis were not included and will be discussed later in this article.