A prospective study on the microbes isolated from the alimentary tract in 120 bone marrow transplant (BMT) recipients (1991-1993) was undertaken to define the spectrum of organisms isolated under antimicrobial prophylaxis, their temporal sequence of emergence, and the associated morbidity and mortality. Clostridium difficile (n = 20), isolated in the pre-engraftment and early post-engraftment periods (day 2-45 post-BMT), was the most common microbe recovered from stool of patients with diarrhea. In contrast to previous reports, no significant difference in mortality was observed between patients with and without C. difficile isolated in stool. Two patients had neutropenic ileocecitis with concomitant bacteremia due to Escherichia coli and Klebsiella pneumoniae. One patient was found to have astrovirus gastroenteritis (day 7), and Giardia lamblia was recovered from the stool of another (day -7). Heavy growth of Staphylococcus aureus from direct smear-positive specimens was found from the upper airway of two patients with severe mucositis and complete dysphagia (day 12 and 23). Salmonella spp. of groups B and E were found in the stool of five asymptomatic patients at the time of conditioning. No specific organisms was recovered from the endoscopic brushing of two patients with lower end esophagitis, three patients with upper gastrointestinal bleeding, and three patients with perirectal cellulitis. During the post-engraftment period, five patients had documented cytomegalovirus gastroenterocolitis (days 34-97), one had Mycobacterium chelonae colitis (day 70), and another had nodular gastritis due to Acremonium falciforme (day 270). Overall, only 28% of patients with alimentary tract symptoms/syndrome had specific pathogens isolated from clinical specimens. Differentiation of the causation of alimentary tract symptoms was often difficult because noninfectious complications such as conditioning toxicity, graft-versus-host disease, and its treatment often caused alimentary tract symptoms in addition to predisposed BMT patient to infection. The reluctance of obtaining tissue biopsy for ascertaining the importance of those potential alimentary tract pathogens often dictate the use of empirical treatment.