Efficacy of a symbiotic product during clinical relapse of ulcerative colitis.

persistently negative and there was no disease recurrence. Thus, the recovery from infection and C. coli clearance could not been achieved, despite several courses of antibiotics, until an attempt to sterilize the gastrointestinal tract with bacitracin/neomicyn was done. The high frequency of Campylobacter found in symptomatic and asymptomatic patients with PAD might be the result of failure to eradicate bacteria, causing a condition of intestinal carriage. It has been showed that the failure of humoral immune response might permit the colonization of epithelium and lamina propria with a possible induction of tissue damage. In this scenario, the observed susceptibility to extracellular bacterial infections such as Helicobacter and Campylobacter indicates the essential role of humoral immunity. The inability to mount an antibody response might account for the genetic stability of the microorganism over time and for the high frequency of C. coli carriage in patients with antibody defects. A patient with XLA with cellulitis associated with C. coli bacteremia relapses after a disease-free interval of more than 5 years has been already described in 2004; in this patient, the pathogen isolated from the blood was considered responsible for the cellulitis, although C. coli was not directly obtained from the skin lesion. What was unexpected in our patient with XLA was the shift from a silent infection to a life-threatening disease, a shift not caused by a change in the molecular pattern of the C. coli strain isolated from stools, skin lesion, and blood. The prolonged and relapsing nature of the life-threatening infection could not be attributed to microbial and/or host genetic factors. Thus, environmental factors should be taken in consideration. The recent description of a patient with XLA with a Campylobacter isolated from intestinal mucosal biopsies, who presented bacteremia in the absence of diarrhoea and with negative stool cultures, prompted us to hypothesize in our patient a sudden break in an intestinal site; however, the colonoscopy did not show any mucosal alteration. Whatever is the underlying pathogenetic mechanism, the high prevalence of C. coli intestinal carriage in PAD suggests to introduce periodic stool cultures with high sensitive techniques and the treatment of positive symptomatic and asymptomatic patients with nonabsorbable antibiotics to prevent a rare, but possible, lifethreatening infection.