Antibiotics as a First-line Therapy for Crohnʼs Disease

Consensus is something you reach when you can't agree. Antibiotics have been described as one of the great “underground” therapies for Crohn's disease. 1 This implies widespread, poorly documented usage despite weak or lacking evidence for efficacy. Perhaps this is so, although in view of the multitude of biologic and other agents in clinical trials worldwide, the level of deployment of antibiotics as a first-line therapy in most tertiary care centers varies. The robust enthusiasm for antibiotics expressed by some clinicians is at variance with the remarkable paucity of controlled clinical trials in the area during the past five decades. Indeed, the scarcity of data seems to be the single area of agreement in the opposing opinion pieces expressed by Drs. Greenberg and Wild in this issue of the journal. Though lack of interest from the pharmaceutical sector might account for the relative paucity of effort applied to controlled clinical trials of antibiotics in Crohn's disease, 2 there must also be a sneaking suspicion that many clinician investigators lack motivation and true enthusiasm for the proposition. Are clinician investigators a different group than the inflammatory bowel disease (IBD) clinicians asserted by Dr. Greenberg to “attest to the marked benefit of antibiotic therapy in individual patients”? The old dictum “those who have enthusiasm have no controls and those who have controls have no enthusiasm” seems to apply to antibiotic therapy in Crohn's disease. Whether IBD clinicians are enthusiastic about antibiotics for treating luminal Crohn's disease, they have enthusiastically (and without rigorous evidence) embraced the use of these agents in perianal Crohn's disease. While two veteran IBD clinicians argue the merits of antibiotic therapy in the pages of this journal, individual IBD clinicians (like Dr. Wild) avoid antibiotics for treating luminal disease on the basis of lack of evidence but advocate for antibiotics for perianal disease, despite the lack of evidence. Notwithstanding the wealth of evidence in support of immunomodifiers and other forms of non-antibiotic therapy for Crohn's disease, it must be conceded that comprehensive management cannot be based on evidence alone. There remain numerous clinical aspects of patient care that have not yet been addressed in clinical trials. This means that clinicians must base their therapeutic strategy upon considerations such as: Is there a sound rationale or biologic plausibility for the use of antibiotics in Crohn's disease? What is the quality of existing evidence for efficacy? How significant is the risk of broad-spectrum antibiotic usage as a first-line strategy? What are the alternatives? The notion that antibiotics are a suitable first-line therapy for Crohn's disease has gained some credence with the popularization of the concept that the pathogenesis involves a convergence of genes, T cells, and bacteria. In common with most chronic inflammatory disorders, a body of evidence favors the triad of genetic susceptibility, environmental triggers or modifiers, and immune-mediated tissue injury. 3 An inappropriate response to normal indigenous flora in genetically susceptible individuals rather than an appropriate immune response to abnormal flora (or infectious agent) is the most widely held working hypothesis. 4,5 The most persuasive evidence cited in support of this is derived from experimental animal models. A diversity of genetically determined defects either at the level of the immune regulation or at the level of mucosal barrier function predispose to Crohn's-like disease. 6 Irrespective of the underlying defect, the enteric flora is an essential requirement for full clinical expression of the inflammatory disease. This is usually cited as circumstantial support for the plausibility of antibiotic therapy in Crohn's disease. However, one might counter-argue that the flora only has a permissive role rather than a stimulatory effect. The flora is required for immune development and immune activation; without enteric bacteria it is doubtful that the mucosal immune system could mount a vigorous inflammatory response. The occurrence of various antibodies in the serum of patients with Crohn's disease against components of the flora is frequently cited to implicate a causative role for the flora in pathogenesis. 3,4 However, none of these has been shown to cause tissue injury, and most may reflect cross reactivity between bacterial and self-antigens. With regard to the quality of evidence supporting the use of antibiotics in Crohn's disease, few would dispute their role in managing some of the complications of Crohn's disease. Few dispute their role in perianal disease. In other forms of Crohn's disease, there appears to be consensus that antibiotics have little if any influence on small bowel disease, albeit with some impact on Crohn's colitis. 1,2 It also seems reasonable to infer consensus from Greenberg and Wild that the significance and extent of efficacy of antibiotics in colonic disease is unknown at present. Whether broad-spectrum antibiotic usage is a realistic first-line strategy even for disease confined to the colon must be tempered with concerns surrounding the hazards of this strategy. The limited data that are available relates mainly to metronidazole and ciprofloxacin. The side-effect profiles of both of these agents as highlighted by Wild in this issue of the journal are not trivial. At 20 mg/kg, metronidazole may be expected to cause unpleasant side effects, and many patients will find the disulfiram-like reactions with alcohol a significant burden and added impact on quality of life. This does little to inspire confidence in a putative first-line therapy and seems to be reflected in the high drop-out rate in controlled studies with nearly half the study population withdrawing in the largest study. 7 Greenberg himself questions whether metronidazole and ciprofloxacin are the optimal antibiotic combination, yet these are the agents widely used when clinicians decide to use any antibiotics. As pointed out by Greenberg, the best evidence in a controlled trial for an antibiotic are the data showing that ornidazole can decrease recurrence rates at 1 year postoperatively in patients with ileocolonic anastomoses. 8 These are patients that may have little overt residual disease; and the report fits with the animal data where antibiotics are more effective in preventing inflammation than in treating active inflammation. Another scenario where antibiotics would have a plausible role is in patients with bacterial overgrowth. This condition can exaggerate symptoms of bloating, cramps, and diarrhea and should be considered in patients with enteric strictures or fistulas. Because the majority of patients with Crohn's disease will progress to have fibrostenotic or penetrating disease over time, 9 it may be that a role for antibiotics is in treating well established rather than early disease. Curiously, there is greater consensus on the use of antibiotics for perianal disease. Antibiotics in perianal disease is as empiric a choice as the use would be for treating luminal disease. A commonly proposed algorithm for treating perineal disease is to start with antibiotics, and upon failure, progress to infliximab. 10 Is it any less rational to embrace this approach with luminal disease? If one accepts the biologic plausibility of therapeutic manipulation of the gut flora in treating Crohn's disease,11 the question arises whether there is a more appropriate alternative strategy. Long-term antibiotic usage poses the risk of antibiotic resistance and even selective overgrowth with organisms such as Clostridium difficile. Instead of eliminating bacterial populations in a nonspecific fashion, it may be more logical to selectively enhance those subsets of the flora that may compete with proinflammatory organisms or even have a direct anti-inflammatory effect on the mucosal immune system. Probiotics, prebiotics, or even engineered functional microbes (turbo probiotics) promise to achieve this, but none can be regarded as evidence-based first-line treatments yet. 12,–14 One possibility not considered by either the advocate or the opponent of antibiotics in Crohn's disease is the use of probiotics following the administration of broad-spectrum antimicrobials. The latter might open up a microbial niche, which can then be occupied by a non-inflammatory organism (probiotic). Experience with this strategy in ulcerative colitis patients with pouchitis is encouraging. 15,–17 In conclusion, the opposing opinions expressed by Drs. Greenberg and Wild in this issue of the journal concur on two points; there is some rationale for therapeutic modification of the flora, but there is a scarcity of data for antibiotics as the first-line strategy to achieve this. Either we need better antibiotics, better selection of patient subsets, or better alternatives. The way forward is to exchange debate with data. However, in the absence of evidence to guide antibiotic or patient choice at present, there is still room for empirical trials of antibiotics in Crohn's disease. Generally (and compared with newer biologic therapies, in particular) they can be safe, inexpensive, and easily administered. Many IBD clinicians are doing this anyway, and they shouldn't feel guilty about it.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.