No abstract

SUMMARY The effect of purified toxin A and partially purified toxin B on rabbit ileum and colon was investigated. Toxin A caused tissue damage which was followed by permeability changes and fluid accumulation in both tissues. Toxin A did not increase the permeability of the colon to the extent observed for ileum; secreted fluid contained less protein of plasma origin. Toxin B had no effect on either tissue. Secretory a-id tissue damaging properties of crude C difficile toxins were found. to be due to toxin A.Clostridium difficile is now known to be the primary cause of antibiotic associated pseudomembranous colitis.1 2 The organism produces four toxins of possible relevance in the pathofenesis of the disease: an enterotoxin (toxin A),3 a cytotoxin (toxin B),3 4 a motility altering factor5 and a recently described labile toxin which causes clear fluid accumulation in rabbit ileal loops.6 In contrast with cholera and Escherichia coli heat labile toxins which induce fluid accumulation with no tissue damage,7 8 toxin A of C difficile causes fluid accumulation with mucosal damage in several animal models: rabbit ileal loop test,9 10 hamster ileal loop9 and hamster caecal loop.9 10 Toxin A is also lethal to mice and causes increased vascular permeability in guinea pig skin.9 Thus toxin A appears to have a mode of action different to that of cholera or E coli heat labile toxins because it causes tissue damage in combination with fluid secretion. Toxin B does not cause fluid accumulation in rabbit ileal loops9 11 or hamster ileal loops. 12 There are discrepancies, however, as to its tissue damag' -g properties:Taylor et a19 report that only limited histological damage is caused to mucosal epithelium of hamster caeca whereas Libby et al'2 showed substantial damage. Toxin B is cytotoxic for a variety of cell lines in vitro but the relevance of this observation to fluid secretion in the intestine is unclear.

Samples of ileal serosa and mesenteric lymph nodes have been harvested before antibiotic administration during 46 non-contaminated operations for Crohn's disease and compared with 43 operations for conditions other than Crohn's. Potentially pathogenic bacteria were isolated from the serosa in 12 (27 per cent) Crohn's patients, compared with 6 (15 per cent) controls (P = 0.04). Intestinal bacteria were recovered from mesenteric nodes in 15 (33 per cent) Crohn's patients compared with 2 (5 per cent) controls (P = 0.006). These findings suggest that bacteria leak from the small bowel lumen in a high proportion of Crohn's disease patients. This may explain the pathogenesis of abscess and fistula in this disorder as well as the high rate of sepsis following elective surgery even in the absence of macroscopic contamination.

No abstract

No abstract

No abstract

No abstract

We have undertaken a prospective randomized trial of one month's antimicrobial therapy for patients with symptomatic relapse of Crohn's disease. Criteria for entry included two major symptoms: fever, abdominal pain, diarrhea, weight loss, abdominal mass or complications (excluding perianal disease); and two hematologic abnormalities: hemoglobin, ESR, albumin, C reactive protein, iron, or total iron binding capacity. Patients were monitored for the aforementioned clinical and hematologic (hemoglobin, albumin, CRP) parameters over six weeks and for changes in fecal flora. Randomization was to four groups: metronidazole alone (M), cotrimoxazole alone (C), metronidazole and cotrimoxazole (C plus M), or double placebo (P). Seventy-two patients entered the study (18 = M, 16 = C, 21 = C plus M, 17 = P). After two weeks, improvement was reported as follows: M = 67 percent, C = 17 percent, C plus M = 71 percent, P = 35 percent. In the metronidazole group, two patients required surgery and one had troublesome side effects. In the cotrimoxazole group, two had side effects. In the combined group (C plus M), four had troublesome side effects and two of the placebo group (P) required operation. By four weeks, there was no difference in response among the groups: (M = 44 percent, C = 62 percent, C plus M = 57 percent, P = 41 percent). Antimicrobials had no effect on fecal flora or hematologic parameters. These results indicate that antimicrobials have little therapeutic potential for relapse of intestinal Crohn's disease.