Involvement of pathogenic or potentially pathogenic bacteria in the pathogenesis of inflammatory bowel disease has long been suggested because, among other reasons, the inflammatory response resembles that in infectious bowel diseases. Elevated antibody levels to pathogen antigens and a changed metabolic activity of the intestinal microflora have been detected in patients with Crohn's disease. Several studies have revealed a possible etiologic link between intestinal microorganisms and inflammatory bowel disease. Therefore, several therapeutic strategies, including reduction or dilution of bacterial components in the intestine by antibiotics or intestinal lavage, respectively, inactivation of inflammatory bacterial products, and reconstitution of intestinal microflora have been employed, substantiating the idea that dysfunction of the intestinal mucosal barrier and an alteration of bacterial composition contribute to the inflammatory disease. However, the beneficial effect of restoration of the physiologic intestinal microflora in colonic inflammation by exogenous administration of a viable nonpathogenic bacterium has not been investigated before in a placebo-controlled study. Promising results came from the present pilot study in which the nonpathogenic Escherichia coli strain Nissle 1917 was tested for efficacy and tolerance in maintaining remission in patients with colonic Crohn's disease. Application of the physiologic bacteria reduced the risk for relapse and minimized the need for glucocorticoids. Therefore we are convinced that in Crohn's disease parts of the intestinal microflora, including the host's immune response toward indigenous flora or an impairment of the gut flora's metabolic activity are involved in the development or at least in the onset of relapse from remissive of colonic Crohn's disease. However, more data are necessary to prove the benefit of E. coli strain Nissle 1917 as a new therapy to maintain remission of colonic Crohn's disease.