We thank Dr. Yarze for his valuable comments. When we refer to Shiga-like toxins, we are thinking about the syndrome of hemorrhagic colitis linked to verotoxigenic strains of E. coli, most often serotype O157:H7. Because stool culture for these strains was not done and stools were not examined for free toxins, we could not definitively exclude enterohemorrhagic E. coli infection. We therefore reported that we believed our patient's case was caused by B. hominis. However, there are several reasons to consider the etiologic role of B. hominis. First, although cases of infection with E. coli O157:H7 have been reported in the United States, Canada, and Great Britain [1], this infection seems to occur rarely in Spain [2]. Second, transmission of E. coli O157:H7 has been linked almost exclusively to consumption of processed food [3], which was not present in our patient. Third, when enterohemorrhagic infection with E. coli is present, colonoscopic examination shows an increasing frequency and severity of mucosal abnormalities from the rectum to the cecum; the most severe changes appear in the cecum and ascending colon [1]. The transverse and upper-ascending colon appeared normal on macroscopic examination of the patient, a finding not mentioned in our report (the cecum was not explored by endoscopy). Finally, submucosal hemorrhage, edema, and fibrin exudation are the most prominent microscopic features of infection with E. coli O157:H7; ulceration, hemorrhage, and capillary thrombi in the mucosa are less common [1]. Our patient had none of these conditions.

Ischemia was one of the initial diagnostic hypotheses we considered for our patients' proctocolitis. However, we favored an infectious cause because of the rectal involvement seen during colonoscopy; the lack of frequent ischemic colitis (as Dr. Yarze stated); the absence of "thumbprinting," tubular narrowing, "sawtooth" irregularity, and sacculations in the subsequent barium enema; and the result of the pathologic examination (which failed to show ischemic necrosis, ulceration, submucosal inflammation and fibrosis, or thrombosis in small vessels [4]). Moreover, we agree with Dr. Yarze that colonic ischemia cannot be absolutely ruled out and that it should be considered when an older patient presents with symptoms and signs of hemorrhagic colitis.