Crohn’s disease
The majority of this work has been conducted using metronidazole, ciprofloxacin or clarithromycin alone or in combination. The two antibiotics most commonly used in the treatment of CD are metronidazole and ciprofloxacin. These antibiotics are active against two of the bacterial species suspected to be involved in the pathogenesis and, at least, relieve the symptoms of CD. Their use is widely accepted for the treatment of perianal fistula and colitis, although there is incontrovertible evidence supporting their use for these conditions.
The few randomized controlled trials to study metronidazole and/or ciprofloxacin have mostly presented negative results (5), although the treatment has been reported to be more effective in those patients whose disease involved the colon (6,7).
One study made use of rifaximin, a non-absorbed rifamycin drug with a broad spectrum of activity against Gram-positive, Gram-negative and even colonic anaerobic bacteria. Owing to the fact that it is not absorbed, systemic adverse effects are rare. In a double-blind, placebo-controlled trial reported by Prantera et al (8), 83 CD patients were divided into 3 groups: group A received rifaximin, 800 mg once a day; group B received rifaximin, 400 mg twice daily; and group C received placebo. Remission rates were significantly higher (52%) in group B than in group A (32%) and in the placebo group (33%) (8). Thus, the antibiotics with cell activity, including the macrolide antibiotics azithromycin and clarithromycin, may provide a more effective chemotherapy. Based on the role of intestinal microflora in the pathogenesis of CD, the use of antibiotics or combined therapy, appears to be a rational strategy. In a previously published study, ciprofloxacin was shown to be effective when added to the existing medium-active treatment in therapy-resistant CD patients (9). Ciprofloxacin, a quinolone drug, is clinically effective in preventing the growth of intestinal bacteria and appears to have immunomodulatory properties. Accumulating evidence supports its role as the primary treatment for CD. For example, in a prospective study by Colombel et al (16), ciprofloxacin (500 mg twice daily) was as effective as mesalazine (4 g/day) and induced remission in patients with mild to moderate CD.
Two important observations have strengthened the bacterial hypothesis for CD. Firstly, genetic studies have identified mutations in the NOD2 gene and the IRGM and ATG16L1 autophagy genes (17). These mutations mean that cells do not contain intracellular bacterial replication, and may have defects in their ability to eliminate bacteria (18). Secondly, an increased number of mucosa-associated E. coli have been identified in patients with IBD (19).
In addition to additional data being required to confirm the effectiveness of antibiotics in the treatment of CD, there is the emergence of resistant bacterial strains and the possible infection by Clostridium difficile to be considered. A study has reported the previous use of antibiotics by the majority of a group of IBD patients with Clostridium difficile infection (20). In an uncontrolled trial, antibiotic therapy resulted in clinical remission in 49 cases (68%) (21). In the same study, 55 patients (76%) showed a clinical response, which was higher in those patients who were also taking steroids (26/29) than those individuals who were not (29/43).
A large amount of data support the use of antibiotics in patients with CD and with diverticulitis. Antibiotics may act via different pathways in patients with CD. They may inhibit the bacteria linked to the pathological processes of the disease or simply lower the luminal bacterial overgrowth. The suppression of intestinal flora may reduce the strength of certain symptoms, including pain and diarrhea.
In our analysis, CD patients undergoing antibacterial therapy are 1.35-fold more likely to experience clinical remission than patients who are not undergoing antibacterial therapy. Homogeneity among the trials was established based on a graphic display and statistical tests. In addition, the meta-analysis of short-term antimicrobial use also revealed that treatment with antibiotics was effective in the induction of clinical remission. These results were clinically and statistically significant.
Whether antibiotics are of actual benefit remains to be determined since only a few controlled trials have been completed. Many used a small number of volunteers or used subgroups which were not always clearly defined. Antibiotic use has predominantly involved metronidazole or ciprofloxacin, and many patients cannot tolerate these drugs for a long duration. Therefore, these antibiotics are used in the treatment of ileal, ileocolonic and colonic CD (22).
Ulcerative colitis
Several different antibiotics, alone and in combination, have been evaluated in the primary and adjuvant treatment of active UC. The routine use of antibiotics is not recommended in mild or moderate UC. In a prospective, double-blind, randomized controlled trial reported by Gilat et al (23), oral metronidazole and sulfasalazine were used for the outpatient management of mild to moderate UC. This revealed that metronidazole was significantly less effective than sulfasalazine by endoscopic and clinical criteria.
In a randomized controlled trial reported by Chapman et al (24), 39 patients with severe UC were treated with either intravenous metronidazole (500 mg every 8 h) or placebo as an adjunct to intravenous steroids for 5 days. No significant difference was observed in clinical improvement between the two groups. In two randomized controlled trials reported by Mantzaris et al (25,26), patients with mild to severe acute UC received either oral or intravenous ciprofloxacin for 2 weeks as an adjunct to corticoid therapy. No significant difference in clinical improvement was observed, with 71% (24/34) and 79% (23/29) of the patients in the ciprofloxacin-treated group and 72% (26/36) and 77% (20/26) of the patients in the placebo-treated group achieving remission. The authors concluded that a short course of oral ciprofloxacin treatment did not appear to increase the proportion of patients with active UC going into remission. In another study, the combination of intravenous metronidazole and tobramycin as an adjunct to corticosteroids was not found to be effective in causing clinical improvement compared with placebo after 10 days of therapy in the treatment of severe UC (27).
However, antibiotics may have a certain benefit as adjuncts to standard anti-inflammatory treatment. Burke et al (28) reported a study in which 84 cases of acute UC were randomized to receive oral tobramycin or placebo for 7 days as an adjunct to steroidal therapy. When evaluated 18–21 days after the end of treatment, 31/42 patients (74%) in the tobramycin group and 18/42 patients (43%) in the placebo group had achieved clinical remission (p=0.008).
Despite considerable evidence for the involvement of bacteria in UC, broad-spectrum antibiotics are usually not used for the treatment of this disease; severe cases are the exception. Only a few antibiotics have been used in the treatment of UC, whether individually or in combination. One trial has shown that rifaxamin is beneficial in the treatment of UC (29). Turunen et al (30) performed a randomized, controlled trial which suggested that ciprofloxacin is beneficial as an adjunctive treatment for active UC during the first 6-month period of administration. At 6 months, the treatment-failure rate in the ciprofloxacin-treated group was 21% (8/38 patients), which was significantly lower than the placebo group rate of 44% (20/45 patients). The authors concluded that the use of ciprofloxacin therapy for 6 months in UC improved the results of conventional treatment with mesalazine and prednisone.
In our study, UC patients undergoing antibacterial therapy have been determined to be 2.17-fold more likely to experience clinical remission than patients receiving no antibacterial therapy. Homogeneity among the trials was established based on a graphic display (Fig. 2) and statistical tests. In addition, meta-analysis of short-term antibacterial trials revealed that 5–180 days of antibiotic therapy is effective for clinical remission. These results were clinically and statistical significant.
However, a lack of well-designed, placebo-controlled trials has made the efficacy of antibiotics as the primary treatment for IBD questionable. Poor study design, high drop-out rates and insufficient numbers of subjects in the current studies have led to negative or equivocal results causing further controversy.
To date, these studies have provided sufficient evidence that antibiotics have been useful in the treatment of this disease. The use of antibiotics in patients with IBD, however, is controversial. The treatment of experimental IBD with antibiotics may have contradictory results, since it remains unclear which bacteria cause the disease. Moreover, many organisms which grow in the gut mucosa, where they are more resistant to standard antibiotic treatment, may not be able to reach the mucosal surface. Susceptibility testing of organisms has not been carried out in a systematic manner, and antibiotic treatment is not aimed at specific bacteria. In fact, other clinical infections may be present. It would be a difficult and time-consuming task to complete in clinical practice if multiple varieties of bacteria were involved. This is due to the requirement for the mucosal microbial flora of each patient to be characterized and a drug sensitivity test on each patient to be carried out.
Therefore, to determine the effect of antibiotics in the treatment of IBD, larger randomized clinical trials of antibiotics should be carried out, either in solo or combined with other antibiotics or therapies.